bump @llamaindex/server

fix llamacloud api and markdown issue
2026-07-15 05:08:15 -04:00 · 2025-04-02 17:20:32 +07:00 · 2025-04-02 16:51:21 +07:00
616 changed files with 9102 additions and 34937 deletions
@@ -0,0 +1,5 @@
+---
+"create-llama": patch
+---
+
+fix: add trycatch for generating error
@@ -1,6 +0,0 @@
---
-"create-llama": patch
-"@llamaindex/server": patch
---
-
-fix: unsupported language in code gen workflow
@@ -0,0 +1,5 @@
+---
+"create-llama": patch
+---
+
+bump: chat-ui and tailwind v4
@@ -0,0 +1,12 @@
+{
+  "extends": [
+    "prettier"
+  ],
+  "rules": {
+    "max-params": [
+      "error",
+      4
+    ],
+    "prefer-const": "error",
+  },
+}
@@ -1,15 +1,12 @@
-name: E2E Tests for create-llama package
+name: E2E Tests
 on:
  push:
    branches: [main]
-    paths-ignore:
-      - "python/llama-index-server/**"
-      - ".github/workflows/*llama_index_server.yml"
  pull_request:
    branches: [main]
-    paths-ignore:
-      - "python/llama-index-server/**"
-      - ".github/workflows/*llama_index_server.yml"
+
+env:
+  POETRY_VERSION: "1.6.1"

 jobs:
  e2e-python:
@@ -23,7 +20,6 @@ jobs:
        os: [macos-latest, windows-latest, ubuntu-22.04]
        frameworks: ["fastapi"]
        datasources: ["--no-files", "--example-file", "--llamacloud"]
-        template-types: ["streaming", "llamaindexserver"]
    defaults:
      run:
        shell: bash
@@ -36,10 +32,10 @@ jobs:
        with:
          python-version: ${{ matrix.python-version }}

-      - name: Install uv
-        run: curl -LsSf https://astral.sh/uv/install.sh | sh
-      - name: Add uv to PATH # Ensure uv is available in subsequent steps
-        run: echo "$HOME/.cargo/bin" >> $GITHUB_PATH
+      - name: Install Poetry
+        uses: snok/install-poetry@v1
+        with:
+          version: ${{ env.POETRY_VERSION }}

      - uses: pnpm/action-setup@v3

@@ -54,15 +50,15 @@ jobs:

      - name: Install Playwright Browsers
        run: pnpm exec playwright install --with-deps
-        working-directory: packages/create-llama
+        working-directory: .

      - name: Build create-llama
        run: pnpm run build
-        working-directory: packages/create-llama
+        working-directory: .

      - name: Install
        run: pnpm run pack-install
-        working-directory: packages/create-llama
+        working-directory: .

      - name: Run Playwright tests for Python
        run: pnpm run e2e:python
@@ -71,16 +67,13 @@ jobs:
          LLAMA_CLOUD_API_KEY: ${{ secrets.LLAMA_CLOUD_API_KEY }}
          FRAMEWORK: ${{ matrix.frameworks }}
          DATASOURCE: ${{ matrix.datasources }}
-          TEMPLATE_TYPE: ${{ matrix.template-types }}
-          PYTHONIOENCODING: utf-8
-          PYTHONLEGACYWINDOWSSTDIO: utf-8
-        working-directory: packages/create-llama
+        working-directory: .

      - uses: actions/upload-artifact@v4
        if: always()
        with:
-          name: playwright-report-python-${{ matrix.os }}-${{ matrix.frameworks }}-${{ matrix.datasources }}-${{ matrix.template-types }}
-          path: packages/create-llama/playwright-report/
+          name: playwright-report-python-${{ matrix.os }}-${{ matrix.frameworks }}-${{ matrix.datasources }}
+          path: ./playwright-report/
          overwrite: true
          retention-days: 30

@@ -90,12 +83,11 @@ jobs:
    strategy:
      fail-fast: true
      matrix:
-        node-version: [20, 22]
+        node-version: [18, 20]
        python-version: ["3.11"]
        os: [macos-latest, windows-latest, ubuntu-22.04]
-        frameworks: ["nextjs"]
+        frameworks: ["nextjs", "express"]
        datasources: ["--no-files", "--example-file", "--llamacloud"]
-        template-types: ["streaming", "llamaindexserver"]
    defaults:
      run:
        shell: bash
@@ -108,10 +100,10 @@ jobs:
        with:
          python-version: ${{ matrix.python-version }}

-      - name: Install uv
-        run: curl -LsSf https://astral.sh/uv/install.sh | sh
-      - name: Add uv to PATH # Ensure uv is available in subsequent steps
-        run: echo "$HOME/.cargo/bin" >> $GITHUB_PATH
+      - name: Install Poetry
+        uses: snok/install-poetry@v1
+        with:
+          version: ${{ env.POETRY_VERSION }}

      - uses: pnpm/action-setup@v3

@@ -126,15 +118,15 @@ jobs:

      - name: Install Playwright Browsers
        run: pnpm exec playwright install --with-deps
-        working-directory: packages/create-llama
+        working-directory: .

      - name: Build create-llama
        run: pnpm run build
-        working-directory: packages/create-llama
+        working-directory: .

      - name: Install
        run: pnpm run pack-install
-        working-directory: packages/create-llama
+        working-directory: .

      - name: Run Playwright tests for TypeScript
        run: pnpm run e2e:typescript
@@ -143,13 +135,12 @@ jobs:
          LLAMA_CLOUD_API_KEY: ${{ secrets.LLAMA_CLOUD_API_KEY }}
          FRAMEWORK: ${{ matrix.frameworks }}
          DATASOURCE: ${{ matrix.datasources }}
-          TEMPLATE_TYPE: ${{ matrix.template-types }}
-        working-directory: packages/create-llama
+        working-directory: .

      - uses: actions/upload-artifact@v4
        if: always()
        with:
-          name: playwright-report-typescript-${{ matrix.os }}-${{ matrix.frameworks }}-${{ matrix.datasources }}-node${{ matrix.node-version }}-${{ matrix.template-types }}
-          path: packages/create-llama/playwright-report/
+          name: playwright-report-typescript-${{ matrix.os }}-${{ matrix.frameworks }}-${{ matrix.datasources }}-node${{ matrix.node-version }}
+          path: ./playwright-report/
          overwrite: true
          retention-days: 30
@@ -31,21 +31,12 @@ jobs:
      - name: Run Prettier
        run: pnpm run format

-      - name: Run build
-        run: pnpm run build
-
-      - name: Run Typecheck for examples
-        run: pnpm run typecheck
-        working-directory: packages/server/examples
-
      - name: Run Python format check
        uses: chartboost/ruff-action@v1
        with:
          args: "format --check"
-          src: "python/llama-index-server"

      - name: Run Python lint
        uses: chartboost/ruff-action@v1
        with:
          args: "check"
-          src: "python/llama-index-server"
@@ -5,7 +5,7 @@ on:
    branches:
      - main
    paths:
-      - "python/llama-index-server/**"
+      - "llama-index-server/**"
      - ".github/workflows/release_llama_index_server.yml"
  pull_request:
    types:
@@ -19,11 +19,10 @@ jobs:
    runs-on: ubuntu-latest
    defaults:
      run:
-        working-directory: ./python/llama-index-server
+        working-directory: ./llama-index-server
    if: |
      github.event_name == 'push' && 
-      !startsWith(github.ref, 'refs/heads/release/llama-index-server-v') &&
-      !contains(github.event.head_commit.message, 'Release: llama-index-server v')
+      !startsWith(github.ref, 'refs/heads/release/llama-index-server-v')

    steps:
      - name: Checkout Repository
@@ -31,19 +30,17 @@ jobs:
        with:
          fetch-depth: 0

-      - name: Install uv
-        uses: astral-sh/setup-uv@v5
-        with:
-          enable-cache: true
-
      - name: Set up Python
        uses: actions/setup-python@v5
        with:
          python-version: "3.11"

+      - name: Install Poetry
+        run: |
+          curl -sSL https://install.python-poetry.org | python3 -
+
      - name: Install dependencies
-        shell: bash
-        run: uv sync --all-extras --dev
+        run: poetry install

      - name: Setup Git
        run: |
@@ -51,17 +48,15 @@ jobs:
          git config --global user.name "github-actions[bot]"

      - name: Bump patch version
-        shell: bash
        run: |
-          uvx --from=toml-cli toml set --toml-path=pyproject.toml project.version $(uvx --from=toml-cli toml get --toml-path=pyproject.toml project.version | awk -F. '{$NF = $NF + 1;}1' OFS=.)
+          poetry version patch
          git add pyproject.toml
-          git commit -m "chore(release): bump llama-index-server version to $(uvx --from=toml-cli toml get --toml-path=pyproject.toml project.version)"
+          git commit -m "chore(release): bump version to $(poetry version -s)"

      - name: Get current version
        id: get_version
-        shell: bash
        run: |
-          version=$(uvx --from=toml-cli toml get --toml-path=pyproject.toml project.version)
+          version=$(poetry version -s)
          echo "current_version=${version}" >> "$GITHUB_OUTPUT"

      - name: Create Release PR
@@ -85,7 +80,7 @@ jobs:
    runs-on: ubuntu-latest
    defaults:
      run:
-        working-directory: ./python/llama-index-server
+        working-directory: ./llama-index-server
    if: |
      github.event_name == 'pull_request' && 
      github.event.pull_request.merged == true && 
@@ -96,34 +91,31 @@ jobs:
      - name: Checkout Repository
        uses: actions/checkout@v4

-      - name: Install uv
-        uses: astral-sh/setup-uv@v5
-        with:
-          enable-cache: true
-
      - name: Set up Python
        uses: actions/setup-python@v5
        with:
          python-version: "3.11"

+      - name: Install Poetry
+        run: |
+          curl -sSL https://install.python-poetry.org | python3 -
+
      - name: Install dependencies
-        shell: bash
-        run: uv sync --all-extras
+        run: poetry install

      - name: Get current version
        id: get_version
-        shell: bash
        run: |
-          version=$(uvx --from=toml-cli toml get --toml-path=pyproject.toml project.version)
+          version=$(poetry version -s)
          echo "current_version=${version}" >> "$GITHUB_OUTPUT"

-      - name: Build package
-        shell: bash
-        run: uv build --no-sources
-
-      - name: Publish to PyPI
-        shell: bash
-        run: uv publish --token ${{ secrets.PYPI_TOKEN }}
+      - name: Build and publish to PyPI
+        uses: JRubics/poetry-publish@v2.1
+        with:
+          python_version: "3.11"
+          pypi_token: ${{ secrets.PYPI_TOKEN }}
+          package_directory: "llama-index-server"
+          poetry_install_options: "--without dev"

      - name: Create GitHub Release
        uses: softprops/action-gh-release@v2
@@ -4,6 +4,7 @@ on:
  pull_request:

 env:
+  POETRY_VERSION: "1.8.3"
  PYTHON_VERSION: "3.9"

 jobs:
@@ -12,7 +13,7 @@ jobs:
    runs-on: ${{ matrix.os }}
    defaults:
      run:
-        working-directory: python/llama-index-server
+        working-directory: llama-index-server
    strategy:
      matrix:
        os: [ubuntu-latest, windows-latest]
@@ -20,80 +21,91 @@ jobs:
    steps:
      - uses: actions/checkout@v4

-      - name: Install uv
-        uses: astral-sh/setup-uv@v5
-        with:
-          enable-cache: true
+      - name: Install Poetry
+        run: pipx install poetry==${{ env.POETRY_VERSION }}

-      - name: Set up Python ${{ matrix.python-version }}
+      - name: Set up python ${{ matrix.python-version }}
        uses: actions/setup-python@v5
        with:
          python-version: ${{ matrix.python-version }}
+          cache: "poetry"
+
+      - name: Configure Poetry
+        run: |
+          poetry config virtualenvs.create true
+          poetry config virtualenvs.in-project true
+          poetry env use python

      - name: Install dependencies
        shell: bash
-        run: uv sync --all-extras --dev
+        run: poetry install --with dev

      - name: Run unit tests
        shell: bash
-        run: uv run pytest tests
+        run: |
+          poetry run pytest tests

  type-check:
    name: Type Check
    runs-on: ubuntu-latest
    defaults:
      run:
-        working-directory: python/llama-index-server
+        working-directory: llama-index-server
    steps:
      - uses: actions/checkout@v4

-      - name: Install uv
-        uses: astral-sh/setup-uv@v5
-        with:
-          enable-cache: true
+      - name: Install Poetry
+        run: pipx install poetry==${{ env.POETRY_VERSION }}

      - name: Set up Python
        uses: actions/setup-python@v5
        with:
          python-version: ${{ env.PYTHON_VERSION }}
+          cache: "poetry"
+
+      - name: Configure Poetry
+        run: |
+          poetry config virtualenvs.create true
+          poetry config virtualenvs.in-project true
+          poetry env use python

      - name: Install dependencies
        shell: bash
-        run: uv sync --all-extras --dev
+        run: poetry install --with dev

      - name: Run mypy
        shell: bash
-        run: uv run mypy llama_index
+        run: poetry run mypy llama_index

  build:
    needs: [unit-test, type-check]
    runs-on: ubuntu-latest
    defaults:
      run:
-        working-directory: python/llama-index-server
+        working-directory: llama-index-server
    steps:
      - uses: actions/checkout@v4
-
-      - name: Install uv
-        uses: astral-sh/setup-uv@v5
-        with:
-          enable-cache: true
-
+      - name: Install Poetry
+        run: pipx install poetry==${{ env.POETRY_VERSION }}
      - name: Set up Python
        uses: actions/setup-python@v5
        with:
          python-version: ${{ env.PYTHON_VERSION }}
-
-      - name: Install build package
+      - name: Clear python cache
        shell: bash
-        run: uv sync --all-extras
-
+        run: poetry cache clear --all pypi
+      - name: Build package
+        shell: bash
+        run: poetry build
+      - name: Test installing built package
+        shell: bash
+        run: python -m pip install .
      - name: Test import
        shell: bash
-        run: uv run python -c "from llama_index.server import LlamaIndexServer"
-
+        working-directory: ${{ vars.RUNNER_TEMP }}
+        run: python -c "from llama_index.server import LlamaIndexServer"
      - name: Upload artifact
        uses: actions/upload-artifact@v4
        with:
          name: llama-index-server
-          path: python/llama-index-server/dist/
+          path: llama-index-server/dist/
@@ -6,6 +6,10 @@ node_modules
 .pnpm-store
 .pnp.js

+# testing
+coverage
+.coverage
+
 # next.js
 .next/
 out/
@@ -31,9 +35,31 @@ yarn-error.log*
 dist/
 lib/

+# e2e
+.cache
+test-results/
+playwright-report/
+blob-report/
+playwright/.cache/
+.tsbuildinfo
+e2e/cache
+
 # intellij
 **/.idea

+# Python
+.mypy_cache/
+venv/
+.venv/
+dist/
+.__pycache__
+__pycache__
+.python-version
+.ui
+
+# build artifacts
+create-llama-*.tgz
+
 # vscode
 .vscode
 !.vscode/settings.json
@@ -1,4 +1,3 @@
 pnpm format
 pnpm lint
-uvx ruff check .
-uvx ruff format . --check 
+uvx ruff format --check templates/
@@ -1,17 +1,6 @@
-node_modules/
+apps/docs/i18n
+apps/docs/docs/api
 pnpm-lock.yaml
 lib/
 dist/
-cache/
-build/
-.next/
-out/
-packages/server/server/
-**/playwright-report/
-**/test-results/
-
-# Python
-python/
-**/*.mypy_cache/**
-**/*.venv/**
-**/*.ruff_cache/**
+.docusaurus/
@@ -1,114 +1,5 @@
 # create-llama

-## 0.5.15
-
-### Patch Changes
-
- 527075c: Enable dev mode that allows updating code directly in the UI
-
-## 0.5.14
-
-### Patch Changes
-
- 1df8cfb: Split artifacts use case to document generator and code generator
- 1b5a519: chore: improve dev experience with nodemon
- b3eb0ba: Fix typing check issue
- 556f33c: fix chromadb dependency issue
- 2451539: fix: remove dead generated ai code
- 7a70390: Deprecate pro mode
-
-## 0.5.13
-
-### Patch Changes
-
- f4ca602: Add artifact use case for Typescript template
- f4ca602: Update typescript use cases to use the new workflow engine
-
-## 0.5.12
-
-### Patch Changes
-
- 241d82a: Add artifacts use case (python)
-
-## 0.5.11
-
-### Patch Changes
-
- 3960618: chore: create-llama monorepo
- 8fe5fc2: chore: add llamaindex server package
-
-## 0.5.10
-
-### Patch Changes
-
- 0a2e12a: Use uv as the default package manager
-
-## 0.5.9
-
-### Patch Changes
-
- 4bc53ac: Bump new chat ui and update deep research component
- 4bc53ac: Support generate UI for deep research use case (Typescript)
-
-## 0.5.8
-
-### Patch Changes
-
- 765181a: chore: test typescript e2e with node 20 and 22
-
-## 0.5.7
-
-### Patch Changes
-
- 5988657: chore: bump llmaindex
-
-## 0.5.6
-
-### Patch Changes
-
- d363ced: Bump llamaindex server packages
-
-## 0.5.5
-
-### Patch Changes
-
- ee85320: The default custom deep research component does not work.
-
-## 0.5.4
-
-### Patch Changes
-
- 7c3b279: Support code generation of event components using an LLM (Python)
-
-## 0.5.3
-
-### Patch Changes
-
- 76ec360: Update templates to use new chat ui config
-
-## 0.5.2
-
-### Patch Changes
-
- c9f8f8d: Use custom component for deep research use case
-
-## 0.5.1
-
-### Patch Changes
-
- 08b3e07: Simplify the local index code.
-
-## 0.5.0
-
-### Minor Changes
-
- 54c9e2f: Simplified generated code using LlamaIndexServer
-
-### Patch Changes
-
- 0e4ecfa: fix: add trycatch for generating error
- ee69ce7: bump: chat-ui and tailwind v4
-
 ## 0.4.0

 ### Minor Changes
@@ -55,7 +55,7 @@ Then re-start your app. Remember you'll need to re-run `generate` if you add new
 If you're using the Python backend, you can trigger indexing of your data by calling:

 ```bash
-uv run generate
+poetry run generate
 ```

 ## Customizing the AI models
@@ -106,16 +106,28 @@ Ok to proceed? (y) y
 You can also pass command line arguments to set up a new project
 non-interactively. For a list of the latest options, call `create-llama --help`.

+### Running in pro mode
+
+If you prefer more advanced customization options, you can run `create-llama` in pro mode using the `--pro` flag.
+
+In pro mode, instead of selecting a predefined use case, you'll be prompted to select each technical component of your project. This allows for greater flexibility in customizing your project, including:
+
+- **Vector Store**: Choose from a variety of vector stores for keeping your documents, including MongoDB, Pinecone, Weaviate, Qdrant and Chroma.
+- **Tools**: Choose from a variety of agent tools (functions called by the LLM), such as:
+  - Code Interpreter: Executes Python code in a secure Jupyter notebook environment
+  - Artifact Code Generator: Generates code artifacts that can be run in a sandbox
+  - OpenAPI Action: Facilitates requests to a provided OpenAPI schema
+  - Image Generator: Creates images based on text descriptions
+  - Web Search: Performs web searches to retrieve up-to-date information
+- **Data Sources**: Integrate various data sources into your chat application, including local files, websites, or database-retrieved data.
+- **Backend Options**: Besides using Next.js or FastAPI, you can also select to use Express for a more traditional Node.js application.
+- **Observability**: Choose from a variety of LLM observability tools, including LlamaTrace and Traceloop.
+
+Pro mode is ideal for developers who want fine-grained control over their project's configuration and are comfortable with more technical setup options.
+
 ## LlamaIndex Documentation

 - [TS/JS docs](https://ts.llamaindex.ai/)
 - [Python docs](https://docs.llamaindex.ai/en/stable/)

-## LlamaIndex Server
-
-The generated code is using the LlamaIndex Server, which serves LlamaIndex Workflows and Agent Workflows via an API server. See the following docs for more information:
-
- [LlamaIndex Server For TypeScript](./packages/server/README.md)
- [LlamaIndex Server For Python](./python/llama-index-server/README.md)
-
 Inspired by and adapted from [create-next-app](https://github.com/vercel/next.js/tree/canary/packages/create-next-app)
@@ -1,3 +1,4 @@
+/* eslint-disable import/no-extraneous-dependencies */
 import path from "path";
 import { green, yellow } from "picocolors";
 import { tryGitInit } from "./helpers/git";
@@ -89,7 +90,7 @@ export async function createApp({
  // Install backend
  await installTemplate({ ...args, backend: true });

-  if (frontend && framework === "fastapi" && template !== "llamaindexserver") {
+  if (frontend && framework === "fastapi") {
    // install frontend
    const frontendRoot = path.join(root, ".frontend");
    await makeDir(frontendRoot);
@@ -109,7 +110,7 @@ export async function createApp({
    console.log();
  }

-  if (toolsRequireConfig(tools) && template !== "llamaindexserver") {
+  if (toolsRequireConfig(tools)) {
    const configFile =
      framework === "fastapi" ? "config/tools.yaml" : "config/tools.json";
    console.log(
@@ -0,0 +1,233 @@
+import { expect, test } from "@playwright/test";
+import { exec } from "child_process";
+import fs from "fs";
+import path from "path";
+import util from "util";
+import { TemplateFramework, TemplateVectorDB } from "../../helpers/types";
+import { RunCreateLlamaOptions, createTestDir, runCreateLlama } from "../utils";
+
+const execAsync = util.promisify(exec);
+
+const templateFramework: TemplateFramework = process.env.FRAMEWORK
+  ? (process.env.FRAMEWORK as TemplateFramework)
+  : "fastapi";
+const dataSource: string = process.env.DATASOURCE
+  ? process.env.DATASOURCE
+  : "--example-file";
+
+// TODO: add support for other templates
+
+if (
+  dataSource === "--example-file" // XXX: this test provides its own data source - only trigger it on one data source (usually the CI matrix will trigger multiple data sources)
+) {
+  // vectorDBs, tools, and data source combinations to test
+  const vectorDbs: TemplateVectorDB[] = [
+    "mongo",
+    "pg",
+    "pinecone",
+    "milvus",
+    "astra",
+    "qdrant",
+    "chroma",
+    "weaviate",
+  ];
+
+  const toolOptions = [
+    "wikipedia.WikipediaToolSpec",
+    "google.GoogleSearchToolSpec",
+    "document_generator",
+    "artifact",
+  ];
+
+  const dataSources = [
+    "--example-file",
+    "--web-source https://www.example.com",
+    "--db-source mysql+pymysql://user:pass@localhost:3306/mydb",
+  ];
+
+  const observabilityOptions = ["llamatrace", "traceloop"];
+
+  test.describe("Mypy check", () => {
+    test.describe.configure({ retries: 0 });
+
+    // Test vector databases
+    for (const vectorDb of vectorDbs) {
+      test(`Mypy check for vectorDB: ${vectorDb}`, async () => {
+        const cwd = await createTestDir();
+        const { pyprojectPath } = await createAndCheckLlamaProject({
+          options: {
+            cwd,
+            templateType: "streaming",
+            templateFramework,
+            dataSource: "--example-file",
+            vectorDb,
+            tools: "none",
+            port: 3000,
+            postInstallAction: "none",
+            templateUI: undefined,
+            appType: "--no-frontend",
+            llamaCloudProjectName: undefined,
+            llamaCloudIndexName: undefined,
+            observability: undefined,
+          },
+        });
+
+        const pyprojectContent = fs.readFileSync(pyprojectPath, "utf-8");
+        if (vectorDb !== "none") {
+          if (vectorDb === "pg") {
+            expect(pyprojectContent).toContain(
+              "llama-index-vector-stores-postgres",
+            );
+          } else {
+            expect(pyprojectContent).toContain(
+              `llama-index-vector-stores-${vectorDb}`,
+            );
+          }
+        }
+      });
+    }
+
+    // Test tools
+    for (const tool of toolOptions) {
+      test(`Mypy check for tool: ${tool}`, async () => {
+        const cwd = await createTestDir();
+        const { pyprojectPath } = await createAndCheckLlamaProject({
+          options: {
+            cwd,
+            templateType: "streaming",
+            templateFramework,
+            dataSource: "--example-file",
+            vectorDb: "none",
+            tools: tool,
+            port: 3000,
+            postInstallAction: "none",
+            templateUI: undefined,
+            appType: "--no-frontend",
+            llamaCloudProjectName: undefined,
+            llamaCloudIndexName: undefined,
+            observability: undefined,
+          },
+        });
+
+        const pyprojectContent = fs.readFileSync(pyprojectPath, "utf-8");
+        if (tool === "wikipedia.WikipediaToolSpec") {
+          expect(pyprojectContent).toContain("wikipedia");
+        }
+        if (tool === "google.GoogleSearchToolSpec") {
+          expect(pyprojectContent).toContain("google");
+        }
+      });
+    }
+
+    // Test data sources
+    for (const dataSource of dataSources) {
+      const dataSourceType = dataSource.split(" ")[0];
+      test(`Mypy check for data source: ${dataSourceType}`, async () => {
+        const cwd = await createTestDir();
+        const { pyprojectPath } = await createAndCheckLlamaProject({
+          options: {
+            cwd,
+            templateType: "streaming",
+            templateFramework,
+            dataSource,
+            vectorDb: "none",
+            tools: "none",
+            port: 3000,
+            postInstallAction: "none",
+            templateUI: undefined,
+            appType: "--no-frontend",
+            llamaCloudProjectName: undefined,
+            llamaCloudIndexName: undefined,
+            observability: undefined,
+          },
+        });
+
+        const pyprojectContent = fs.readFileSync(pyprojectPath, "utf-8");
+        if (dataSource.includes("--web-source")) {
+          expect(pyprojectContent).toContain("llama-index-readers-web");
+        }
+        if (dataSource.includes("--db-source")) {
+          expect(pyprojectContent).toContain("llama-index-readers-database");
+        }
+      });
+    }
+
+    // Test observability options
+    for (const observability of observabilityOptions) {
+      test(`Mypy check for observability: ${observability}`, async () => {
+        const cwd = await createTestDir();
+
+        const { pyprojectPath } = await createAndCheckLlamaProject({
+          options: {
+            cwd,
+            templateType: "streaming",
+            templateFramework,
+            dataSource: "--example-file",
+            vectorDb: "none",
+            tools: "none",
+            port: 3000,
+            postInstallAction: "none",
+            templateUI: undefined,
+            appType: "--no-frontend",
+            llamaCloudProjectName: undefined,
+            llamaCloudIndexName: undefined,
+            observability,
+          },
+        });
+      });
+    }
+  });
+}
+
+async function createAndCheckLlamaProject({
+  options,
+}: {
+  options: RunCreateLlamaOptions;
+}): Promise<{ pyprojectPath: string; projectPath: string }> {
+  const result = await runCreateLlama(options);
+  const name = result.projectName;
+  const projectPath = path.join(options.cwd, name);
+
+  // Check if the app folder exists
+  expect(fs.existsSync(projectPath)).toBeTruthy();
+
+  // Check if pyproject.toml exists
+  const pyprojectPath = path.join(projectPath, "pyproject.toml");
+  expect(fs.existsSync(pyprojectPath)).toBeTruthy();
+
+  const env = {
+    ...process.env,
+    POETRY_VIRTUALENVS_IN_PROJECT: "true",
+  };
+
+  // Run poetry install
+  try {
+    const { stdout: installStdout, stderr: installStderr } = await execAsync(
+      "poetry install",
+      { cwd: projectPath, env },
+    );
+    console.log("poetry install stdout:", installStdout);
+    console.error("poetry install stderr:", installStderr);
+  } catch (error) {
+    console.error("Error running poetry install:", error);
+    throw error;
+  }
+
+  // Run poetry run mypy
+  try {
+    const { stdout: mypyStdout, stderr: mypyStderr } = await execAsync(
+      "poetry run mypy .",
+      { cwd: projectPath, env },
+    );
+    console.log("poetry run mypy stdout:", mypyStdout);
+    console.error("poetry run mypy stderr:", mypyStderr);
+  } catch (error) {
+    console.error("Error running mypy:", error);
+    throw error;
+  }
+
+  // If we reach this point without throwing an error, the test passes
+  expect(true).toBeTruthy();
+
+  return { pyprojectPath, projectPath };
+}
@@ -1,3 +1,4 @@
+/* eslint-disable turbo/no-undeclared-env-vars */
 import { expect, test } from "@playwright/test";
 import { ChildProcess } from "child_process";
 import fs from "fs";
@@ -12,30 +13,19 @@ import { createTestDir, runCreateLlama, type AppType } from "../utils";
 const templateFramework: TemplateFramework = process.env.FRAMEWORK
  ? (process.env.FRAMEWORK as TemplateFramework)
  : "fastapi";
-const dataSource: string = process.env.DATASOURCE
-  ? (process.env.DATASOURCE as string)
-  : "--example-file";
-const llamaCloudProjectName = "create-llama";
-const llamaCloudIndexName = "e2e-test";
-
+const dataSource: string = "--example-file";
 const templateUI: TemplateUI = "shadcn";
 const templatePostInstallAction: TemplatePostInstallAction = "runApp";
-const appType: AppType = "--frontend";
+const appType: AppType = templateFramework === "fastapi" ? "--frontend" : "";
 const userMessage = "Write a blog post about physical standards for letters";
-const templateUseCases = [
-  "agentic_rag",
-  "financial_report",
-  "deep_research",
-  "code_generator",
-];
+const templateUseCases = ["financial_report", "blog", "form_filling"];

 for (const useCase of templateUseCases) {
-  test.describe(`Test use case ${useCase} ${templateFramework} ${dataSource} ${templateUI} ${appType} ${templatePostInstallAction}`, async () => {
+  test.describe(`Test multiagent template ${useCase} ${templateFramework} ${dataSource} ${templateUI} ${appType} ${templatePostInstallAction}`, async () => {
    test.skip(
-      dataSource === "--no-files" || templateFramework === "express",
-      "The llamaindexserver template currently only works with nextjs, fastapi. We also only run on Linux to speed up tests.",
+      process.platform !== "linux" || process.env.DATASOURCE === "--no-files",
+      "The multiagent template currently only works with files. We also only run on Linux to speed up tests.",
    );
-    const useLlamaParse = dataSource === "--llamacloud";
    let port: number;
    let cwd: string;
    let name: string;
@@ -48,7 +38,7 @@ for (const useCase of templateUseCases) {
      cwd = await createTestDir();
      const result = await runCreateLlama({
        cwd,
-        templateType: "llamaindexserver",
+        templateType: "multiagent",
        templateFramework,
        dataSource,
        vectorDb,
@@ -57,9 +47,6 @@ for (const useCase of templateUseCases) {
        templateUI,
        appType,
        useCase,
-        llamaCloudProjectName,
-        llamaCloudIndexName,
-        useLlamaParse,
      });
      name = result.projectName;
      appProcess = result.appProcess;
@@ -76,9 +63,7 @@ for (const useCase of templateUseCases) {
          templateFramework === "express",
      );
      await page.goto(`http://localhost:${port}`);
-      await expect(page.getByText("Built by LlamaIndex")).toBeVisible({
-        timeout: 5 * 60 * 1000,
-      });
+      await expect(page.getByText("Built by LlamaIndex")).toBeVisible();
    });

    test("Frontend should be able to submit a message and receive the start of a streamed response", async ({
@@ -87,9 +72,9 @@ for (const useCase of templateUseCases) {
      test.skip(
        templatePostInstallAction !== "runApp" ||
          useCase === "financial_report" ||
-          useCase === "deep_research" ||
+          useCase === "form_filling" ||
          templateFramework === "express",
-        "Skip chat tests for financial report and deep research.",
+        "Skip chat tests for financial report and form filling.",
      );
      await page.goto(`http://localhost:${port}`);
      await page.fill("form textarea", userMessage);
@@ -101,12 +86,6 @@ for (const useCase of templateUseCases) {
      await page.click("form button[type=submit]");

      const response = await responsePromise;
-      console.log(`Response status: ${response.status()}`);
-      const responseBody = await response
-        .text()
-        .catch((e) => `Error reading body: ${e}`);
-      console.log(`Response body: ${responseBody}`);
-
      expect(response.ok()).toBeTruthy();
    });

@@ -1,3 +1,4 @@
+/* eslint-disable turbo/no-undeclared-env-vars */
 import { expect, test } from "@playwright/test";
 import { ChildProcess } from "child_process";
 import fs from "fs";
@@ -1,3 +1,4 @@
+/* eslint-disable turbo/no-undeclared-env-vars */
 import { expect, test } from "@playwright/test";
 import { ChildProcess } from "child_process";
 import fs from "fs";
@@ -0,0 +1,105 @@
+import { expect, test } from "@playwright/test";
+import { exec } from "child_process";
+import fs from "fs";
+import path from "path";
+import util from "util";
+import { TemplateFramework, TemplateVectorDB } from "../../helpers/types";
+import { createTestDir, runCreateLlama } from "../utils";
+
+const execAsync = util.promisify(exec);
+
+const templateFramework: TemplateFramework = process.env.FRAMEWORK
+  ? (process.env.FRAMEWORK as TemplateFramework)
+  : "nextjs";
+const dataSource: string = process.env.DATASOURCE
+  ? process.env.DATASOURCE
+  : "--example-file";
+
+// vectorDBs combinations to test
+const vectorDbs: TemplateVectorDB[] = [
+  "mongo",
+  "pg",
+  "qdrant",
+  "pinecone",
+  "milvus",
+  "astra",
+  "chroma",
+  "llamacloud",
+  "weaviate",
+];
+
+test.describe("Test resolve TS dependencies", () => {
+  // Test vector DBs without LlamaParse
+  for (const vectorDb of vectorDbs) {
+    const optionDescription = `vectorDb: ${vectorDb}, dataSource: ${dataSource}`;
+
+    test(`Vector DB test - ${optionDescription}`, async () => {
+      await runTest(vectorDb, false);
+    });
+  }
+
+  // Test LlamaParse with vectorDB 'none'
+  test(`LlamaParse test - vectorDb: none, dataSource: ${dataSource}, llamaParse: true`, async () => {
+    await runTest("none", true);
+  });
+
+  async function runTest(
+    vectorDb: TemplateVectorDB | "none",
+    useLlamaParse: boolean,
+  ) {
+    const cwd = await createTestDir();
+
+    const result = await runCreateLlama({
+      cwd: cwd,
+      templateType: "streaming",
+      templateFramework: templateFramework,
+      dataSource: dataSource,
+      vectorDb: vectorDb,
+      port: 3000,
+      postInstallAction: "none",
+      templateUI: undefined,
+      appType: templateFramework === "nextjs" ? "" : "--no-frontend",
+      llamaCloudProjectName: undefined,
+      llamaCloudIndexName: undefined,
+      tools: undefined,
+      useLlamaParse: useLlamaParse,
+    });
+    const name = result.projectName;
+
+    // Check if the app folder exists
+    const appDir = path.join(cwd, name);
+    const dirExists = fs.existsSync(appDir);
+    expect(dirExists).toBeTruthy();
+
+    // Install dependencies using pnpm
+    try {
+      const { stderr: installStderr } = await execAsync(
+        "pnpm install --prefer-offline",
+        {
+          cwd: appDir,
+        },
+      );
+    } catch (error) {
+      console.error("Error installing dependencies:", error);
+      throw error;
+    }
+
+    // Run tsc type check and capture the output
+    try {
+      const { stdout, stderr } = await execAsync(
+        "pnpm exec tsc -b --diagnostics",
+        {
+          cwd: appDir,
+        },
+      );
+      // Check if there's any error output
+      expect(stderr).toBeFalsy();
+
+      // Log the stdout for debugging purposes
+      console.log("TypeScript type-check output:", stdout);
+    } catch (error) {
+      console.error("Error running tsc:", error);
+      throw error;
+    }
+  }
+});
@@ -67,8 +67,8 @@ export async function runCreateLlama({
  ].join("-");

  // Handle different data source types
-  const dataSourceArgs = [];
-  if (dataSource.includes("--web-source")) {
+  let dataSourceArgs = [];
+  if (dataSource.includes("--web-source" || "--db-source")) {
    const webSource = dataSource.split(" ")[1];
    dataSourceArgs.push("--web-source", webSource);
  } else if (dataSource.includes("--db-source")) {
@@ -113,12 +113,7 @@ export async function runCreateLlama({
  if (observability) {
    commandArgs.push("--observability", observability);
  }
-  if (
-    (templateType === "multiagent" ||
-      templateType === "reflex" ||
-      templateType === "llamaindexserver") &&
-    useCase
-  ) {
+  if ((templateType === "multiagent" || templateType === "reflex") && useCase) {
    commandArgs.push("--use-case", useCase);
  }

@@ -1,62 +0,0 @@
-import eslint from "@eslint/js";
-import eslintConfigPrettier from "eslint-config-prettier";
-import globals from "globals";
-import tseslint from "typescript-eslint";
-
-export default tseslint.config(
-  eslint.configs.recommended,
-  ...tseslint.configs.recommended,
-  eslintConfigPrettier,
-  {
-    languageOptions: {
-      ecmaVersion: 2022,
-      sourceType: "module",
-      globals: {
-        ...globals.browser,
-        ...globals.node,
-      },
-    },
-  },
-  {
-    files: ["packages/create-llama/**"],
-    rules: {
-      "max-params": ["error", 4],
-      "prefer-const": "error",
-      "no-empty": "off",
-      "no-extra-boolean-cast": "off",
-      "@typescript-eslint/no-explicit-any": "off",
-      "@typescript-eslint/no-unused-vars": "off",
-      "@typescript-eslint/no-empty-object-type": "off",
-      "@typescript-eslint/no-wrapper-object-types": "off",
-      "@typescript-eslint/ban-ts-comment": "off",
-    },
-  },
-  {
-    files: ["packages/server/**"],
-    rules: {
-      "no-irregular-whitespace": "off",
-      "@typescript-eslint/no-unused-vars": "off",
-      "@typescript-eslint/no-explicit-any": [
-        "error",
-        {
-          ignoreRestArgs: true,
-        },
-      ],
-    },
-  },
-  {
-    ignores: [
-      "python/**",
-      "**/*.mypy_cache/**",
-      "**/*.venv/**",
-      "**/*.ruff_cache/**",
-      "**/dist/**",
-      "**/e2e/cache/**",
-      "**/lib/*",
-      "**/.next/**",
-      "**/out/**",
-      "**/node_modules/**",
-      "**/build/**",
-    ],
-  },
-);
@@ -1,3 +1,4 @@
+/* eslint-disable import/no-extraneous-dependencies */
 import { async as glob } from "fast-glob";
 import fs from "fs";
 import path from "path";
@@ -44,7 +44,6 @@ const renderEnvVar = (envVars: EnvVar[]): string => {
 const getVectorDBEnvs = (
  vectorDb?: TemplateVectorDB,
  framework?: TemplateFramework,
-  template?: TemplateType,
 ): EnvVar[] => {
  if (!vectorDb || !framework) {
    return [];
@@ -169,7 +168,7 @@ const getVectorDBEnvs = (
          description:
            "The organization ID for the LlamaCloud project (uses default organization if not specified)",
        },
-        ...(framework === "nextjs" && template !== "llamaindexserver"
+        ...(framework === "nextjs"
          ? // activate index selector per default (not needed for non-NextJS backends as it's handled by createFrontendEnvFile)
            [
              {
@@ -181,7 +180,7 @@ const getVectorDBEnvs = (
            ]
          : []),
      ];
-    case "chroma": {
+    case "chroma":
      const envs = [
        {
          name: "CHROMA_COLLECTION",
@@ -206,7 +205,6 @@ Otherwise, use CHROMA_HOST and CHROMA_PORT config above`,
        });
      }
      return envs;
-    }
    case "weaviate":
      return [
        {
@@ -225,15 +223,13 @@ Otherwise, use CHROMA_HOST and CHROMA_PORT config above`,
        },
      ];
    default:
-      return template !== "llamaindexserver"
-        ? [
-            {
-              name: "STORAGE_CACHE_DIR",
-              description: "The directory to store the local storage cache.",
-              value: ".cache",
-            },
-          ]
-        : [];
+      return [
+        {
+          name: "STORAGE_CACHE_DIR",
+          description: "The directory to store the local storage cache.",
+          value: ".cache",
+        },
+      ];
  }
 };

@@ -386,42 +382,38 @@ const getModelEnvs = (modelConfig: ModelConfig): EnvVar[] => {

 const getFrameworkEnvs = (
  framework: TemplateFramework,
-  template: TemplateType,
  port?: number,
 ): EnvVar[] => {
  const sPort = port?.toString() || "8000";
-  const result: EnvVar[] =
-    template !== "llamaindexserver"
-      ? [
-          {
-            name: "FILESERVER_URL_PREFIX",
-            description:
-              "FILESERVER_URL_PREFIX is the URL prefix of the server storing the images generated by the interpreter.",
-            value:
-              framework === "nextjs"
-                ? // FIXME: if we are using nextjs, port should be 3000
-                  "http://localhost:3000/api/files"
-                : `http://localhost:${sPort}/api/files`,
-          },
-        ]
-      : [];
+  const result: EnvVar[] = [
+    {
+      name: "FILESERVER_URL_PREFIX",
+      description:
+        "FILESERVER_URL_PREFIX is the URL prefix of the server storing the images generated by the interpreter.",
+      value:
+        framework === "nextjs"
+          ? // FIXME: if we are using nextjs, port should be 3000
+            "http://localhost:3000/api/files"
+          : `http://localhost:${sPort}/api/files`,
+    },
+  ];
  if (framework === "fastapi") {
    result.push(
      ...[
        {
          name: "APP_HOST",
-          description: "The address to start the FastAPI app.",
+          description: "The address to start the backend app.",
          value: "0.0.0.0",
        },
        {
          name: "APP_PORT",
-          description: "The port to start the FastAPI app.",
+          description: "The port to start the backend app.",
          value: sPort,
        },
      ],
    );
  }
-  if (framework === "nextjs" && template !== "llamaindexserver") {
+  if (framework === "nextjs") {
    result.push({
      name: "NEXT_PUBLIC_CHAT_API",
      description:
@@ -577,41 +569,25 @@ export const createBackendEnvFile = async (
    | "port"
    | "tools"
    | "observability"
-    | "useLlamaParse"
  >,
 ) => {
  // Init env values
  const envFileName = ".env";
  const envVars: EnvVar[] = [
-    ...(opts.useLlamaParse
-      ? [
-          {
-            name: "LLAMA_CLOUD_API_KEY",
-            description: `The Llama Cloud API key.`,
-            value: opts.llamaCloudKey,
-          },
-        ]
-      : []),
-    ...getVectorDBEnvs(opts.vectorDb, opts.framework, opts.template),
-    ...getToolEnvs(opts.tools),
-    ...getFrameworkEnvs(opts.framework, opts.template, opts.port),
+    {
+      name: "LLAMA_CLOUD_API_KEY",
+      description: `The Llama Cloud API key.`,
+      value: opts.llamaCloudKey,
+    },
    // Add environment variables of each component
-    ...(opts.template === "llamaindexserver"
-      ? [
-          {
-            name: "OPENAI_API_KEY",
-            description: "The OpenAI API key to use.",
-            value: opts.modelConfig.apiKey,
-          },
-        ]
-      : [
-          // don't use this stuff for llama-indexserver
-          ...getModelEnvs(opts.modelConfig),
-          ...getEngineEnvs(),
-          ...getTemplateEnvs(opts.template),
-          ...getObservabilityEnvs(opts.observability),
-          ...getSystemPromptEnv(opts.tools, opts.dataSources, opts.template),
-        ]),
+    ...getModelEnvs(opts.modelConfig),
+    ...getEngineEnvs(),
+    ...getVectorDBEnvs(opts.vectorDb, opts.framework),
+    ...getFrameworkEnvs(opts.framework, opts.port),
+    ...getToolEnvs(opts.tools),
+    ...getTemplateEnvs(opts.template),
+    ...getObservabilityEnvs(opts.observability),
+    ...getSystemPromptEnv(opts.tools, opts.dataSources, opts.template),
  ];
  // Render and write env file
  const content = renderEnvVar(envVars);
@@ -1,3 +1,4 @@
+/* eslint-disable import/no-extraneous-dependencies */
 import { execSync } from "child_process";
 import fs from "fs";
 import path from "path";
@@ -1,7 +1,7 @@
 import { callPackageManager } from "./install";

 import path from "path";
-import picocolors, { cyan } from "picocolors";
+import { cyan } from "picocolors";

 import fsExtra from "fs-extra";
 import { writeLoadersConfig } from "./datasources";
@@ -9,6 +9,7 @@ import { createBackendEnvFile, createFrontendEnvFile } from "./env-variables";
 import { PackageManager } from "./get-pkg-manager";
 import { installLlamapackProject } from "./llama-pack";
 import { makeDir } from "./make-dir";
+import { isHavingPoetryLockFile, tryPoetryRun } from "./poetry";
 import { installPythonTemplate } from "./python";
 import { downloadAndExtractRepo } from "./repo";
 import { ConfigFileType, writeToolsConfig } from "./tools";
@@ -21,7 +22,6 @@ import {
  TemplateVectorDB,
 } from "./types";
 import { installTSTemplate } from "./typescript";
-import { isHavingUvLockFile, tryUvRun } from "./uv";

 const checkForGenerateScript = (
  modelConfig: ModelConfig,
@@ -41,11 +41,7 @@ const checkForGenerateScript = (
    missingSettings.push("your LLAMA_CLOUD_API_KEY");
  }

-  if (
-    vectorDb !== undefined &&
-    vectorDb !== "none" &&
-    vectorDb !== "llamacloud"
-  ) {
+  if (vectorDb !== "none" && vectorDb !== "llamacloud") {
    missingSettings.push("your Vector DB environment variables");
  }

@@ -64,7 +60,7 @@ async function generateContextData(
  if (packageManager) {
    const runGenerate = `${cyan(
      framework === "fastapi"
-        ? "uv run generate"
+        ? "poetry run generate"
        : `${packageManager} run generate`,
    )}`;

@@ -78,21 +74,15 @@ async function generateContextData(
    if (!missingSettings.length) {
      // If all the required environment variables are set, run the generate script
      if (framework === "fastapi") {
-        if (isHavingUvLockFile()) {
+        if (isHavingPoetryLockFile()) {
          console.log(`Running ${runGenerate} to generate the context data.`);
-          const result = tryUvRun("generate");
+          const result = tryPoetryRun("poetry run generate");
          if (!result) {
            console.log(`Failed to run ${runGenerate}.`);
            process.exit(1);
          }
          console.log(`Generated context data`);
          return;
-        } else {
-          console.log(
-            picocolors.yellow(
-              `\nWarning: uv.lock not found. Dependency installation might be incomplete. Skipping context generation.\nIf dependencies were installed, try running '${runGenerate}' manually.\n`,
-            ),
-          );
        }
      } else {
        console.log(`Running ${runGenerate} to generate the context data.`);
@@ -102,14 +92,14 @@ async function generateContextData(
    }

    const settingsMessage = `After setting ${missingSettings.join(" and ")}, run ${runGenerate} to generate the context data.`;
-    console.log(picocolors.yellow(`\n${settingsMessage}\n\n`));
+    console.log(`\n${settingsMessage}\n\n`);
  }
 }

 const downloadFile = async (url: string, destPath: string) => {
  const response = await fetch(url);
  const fileBuffer = await response.arrayBuffer();
-  await fsExtra.writeFile(destPath, new Uint8Array(fileBuffer));
+  await fsExtra.writeFile(destPath, Buffer.from(fileBuffer));
 };

 const prepareContextData = async (
@@ -176,17 +166,6 @@ export const installTemplate = async (

  if (props.framework === "fastapi") {
    await installPythonTemplate(props);
-  } else {
-    await installTSTemplate(props);
-  }
-
-  // write configurations
-  if (props.template !== "llamaindexserver") {
-    await writeToolsConfig(
-      props.root,
-      props.tools,
-      props.framework === "fastapi" ? ConfigFileType.YAML : ConfigFileType.JSON,
-    );
    if (props.vectorDb !== "llamacloud") {
      // write loaders configuration (currently Python only)
      // not needed for LlamaCloud as it has its own loaders
@@ -196,13 +175,26 @@ export const installTemplate = async (
        props.useLlamaParse,
      );
    }
+  } else {
+    await installTSTemplate(props);
  }

+  // write tools configuration
+  await writeToolsConfig(
+    props.root,
+    props.tools,
+    props.framework === "fastapi" ? ConfigFileType.YAML : ConfigFileType.JSON,
+  );
+
  if (props.backend) {
    // This is a backend, so we need to copy the test data and create the env file.

    // Copy the environment file to the target directory.
-    if (props.template !== "community" && props.template !== "llamapack") {
+    if (
+      props.template === "streaming" ||
+      props.template === "multiagent" ||
+      props.template === "reflex"
+    ) {
      await createBackendEnvFile(props.root, props);
    }

@@ -1,3 +1,4 @@
+/* eslint-disable import/no-extraneous-dependencies */
 import spawn from "cross-spawn";
 import { yellow } from "picocolors";
 import type { PackageManager } from "./get-pkg-manager";
@@ -1,3 +1,4 @@
+/* eslint-disable import/no-extraneous-dependencies */
 import fs from "fs";
 import path from "path";
 import { blue, green } from "picocolors";
@@ -143,6 +143,6 @@ export const installLlamapackProject = async ({
  await copyData({ root });
  await installLlamapackExample({ root, llamapack });
  if (postInstallAction === "runApp" || postInstallAction === "dependencies") {
-    installPythonDependencies();
+    installPythonDependencies({ noRoot: true });
  }
 };
@@ -1,3 +1,4 @@
+/* eslint-disable import/no-extraneous-dependencies */
 import { execSync } from "child_process";
 import fs from "fs";

@@ -28,7 +28,7 @@ export async function askModelConfig({
 }: ModelConfigQuestionsParams): Promise<ModelConfig> {
  let modelProvider: ModelProvider = DEFAULT_MODEL_PROVIDER;
  if (askModels) {
-    const choices = [
+    let choices = [
      { title: "OpenAI", value: "openai" },
      { title: "Groq", value: "groq" },
      { title: "Ollama", value: "ollama" },
@@ -3,16 +3,15 @@ import path from "path";
 import { cyan, red } from "picocolors";
 import { parse, stringify } from "smol-toml";
 import terminalLink from "terminal-link";
-import { isUvAvailable, tryUvSync } from "./uv";

 import { assetRelocator, copy } from "./copy";
 import { templatesDir } from "./dir";
+import { isPoetryAvailable, tryPoetryInstall } from "./poetry";
 import { Tool } from "./tools";
 import {
  InstallTemplateArgs,
  ModelConfig,
  TemplateDataSource,
-  TemplateObservability,
  TemplateType,
  TemplateVectorDB,
 } from "./types";
@@ -30,8 +29,6 @@ const getAdditionalDependencies = (
  dataSources?: TemplateDataSource[],
  tools?: Tool[],
  templateType?: TemplateType,
-  observability?: TemplateObservability,
-  // eslint-disable-next-line max-params
 ) => {
  const dependencies: Dependency[] = [];

@@ -40,21 +37,21 @@ const getAdditionalDependencies = (
    case "mongo": {
      dependencies.push({
        name: "llama-index-vector-stores-mongodb",
-        version: ">=0.3.2,<0.4.0",
+        version: "^0.6.0",
      });
      break;
    }
    case "pg": {
      dependencies.push({
        name: "llama-index-vector-stores-postgres",
-        version: ">=0.3.2,<0.4.0",
+        version: "^0.3.2",
      });
      break;
    }
    case "pinecone": {
      dependencies.push({
        name: "llama-index-vector-stores-pinecone",
-        version: ">=0.4.1,<0.5.0",
+        version: "^0.4.1",
        constraints: {
          python: ">=3.11,<3.13",
        },
@@ -64,25 +61,25 @@ const getAdditionalDependencies = (
    case "milvus": {
      dependencies.push({
        name: "llama-index-vector-stores-milvus",
-        version: ">=0.3.0,<0.4.0",
+        version: "^0.3.0",
      });
      dependencies.push({
        name: "pymilvus",
-        version: ">=2.4.4,<3.0.0",
+        version: "2.4.4",
      });
      break;
    }
    case "astra": {
      dependencies.push({
        name: "llama-index-vector-stores-astra-db",
-        version: ">=0.4.0,<0.5.0",
+        version: "^0.4.0",
      });
      break;
    }
    case "qdrant": {
      dependencies.push({
        name: "llama-index-vector-stores-qdrant",
-        version: ">=0.4.0,<0.5.0",
+        version: "^0.4.0",
        constraints: {
          python: ">=3.11,<3.13",
        },
@@ -92,25 +89,21 @@ const getAdditionalDependencies = (
    case "chroma": {
      dependencies.push({
        name: "llama-index-vector-stores-chroma",
-        version: ">=0.4.0,<0.5.0",
-      });
-      dependencies.push({
-        name: "onnxruntime",
-        version: "<1.22.0",
+        version: "^0.4.0",
      });
      break;
    }
    case "weaviate": {
      dependencies.push({
        name: "llama-index-vector-stores-weaviate",
-        version: ">=1.2.3,<2.0.0",
+        version: "^1.2.3",
      });
      break;
    }
    case "llamacloud":
      dependencies.push({
        name: "llama-index-indices-managed-llama-cloud",
-        version: ">=0.6.3,<0.7.0",
+        version: "^0.6.3",
      });
      break;
  }
@@ -123,28 +116,28 @@ const getAdditionalDependencies = (
        case "file":
          dependencies.push({
            name: "docx2txt",
-            version: ">=0.8,<0.9",
+            version: "^0.8",
          });
          break;
        case "web":
          dependencies.push({
            name: "llama-index-readers-web",
-            version: ">=0.3.0,<0.4.0",
+            version: "^0.3.0",
          });
          break;
        case "db":
          dependencies.push({
            name: "llama-index-readers-database",
-            version: ">=0.3.0,<0.4.0",
+            version: "^0.3.0",
          });
          dependencies.push({
            name: "pymysql",
-            version: ">=1.1.0,<2.0.0",
+            version: "^1.1.0",
            extras: ["rsa"],
          });
          dependencies.push({
            name: "psycopg2-binary",
-            version: ">=2.9.9,<3.0.0",
+            version: "^2.9.9",
          });
          break;
      }
@@ -163,122 +156,155 @@ const getAdditionalDependencies = (
    case "ollama":
      dependencies.push({
        name: "llama-index-llms-ollama",
-        version: ">=0.5.0,<0.6.0",
+        version: "0.3.0",
      });
      dependencies.push({
        name: "llama-index-embeddings-ollama",
-        version: ">=0.6.0,<0.7.0",
+        version: "0.3.0",
      });
      break;
    case "openai":
      if (templateType !== "multiagent") {
        dependencies.push({
          name: "llama-index-llms-openai",
-          version: ">=0.3.2,<0.4.0",
+          version: "^0.3.2",
        });
        dependencies.push({
          name: "llama-index-embeddings-openai",
-          version: ">=0.3.1,<0.4.0",
+          version: "^0.3.1",
        });
        dependencies.push({
          name: "llama-index-agent-openai",
-          version: ">=0.4.0,<0.5.0",
+          version: "^0.4.0",
        });
      }
      break;
    case "groq":
+      // Fastembed==0.2.0 does not support python3.13 at the moment
+      // Fixed the python version less than 3.13
+      dependencies.push({
+        name: "python",
+        version: "^3.11,<3.13",
+      });
      dependencies.push({
        name: "llama-index-llms-groq",
-        version: ">=0.3.0,<0.4.0",
+        version: "0.2.0",
      });
      dependencies.push({
        name: "llama-index-embeddings-fastembed",
-        version: ">=0.3.0,<0.4.0",
+        version: "^0.2.0",
      });
      break;
    case "anthropic":
+      // Fastembed==0.2.0 does not support python3.13 at the moment
+      // Fixed the python version less than 3.13
+      dependencies.push({
+        name: "python",
+        version: "^3.11,<3.13",
+      });
      dependencies.push({
        name: "llama-index-llms-anthropic",
-        version: ">=0.6.0,<0.7.0",
+        version: "0.3.0",
      });
      dependencies.push({
        name: "llama-index-embeddings-fastembed",
-        version: ">=0.3.0,<0.4.0",
+        version: "^0.2.0",
      });
      break;
    case "gemini":
      dependencies.push({
        name: "llama-index-llms-gemini",
-        version: ">=0.4.0,<0.5.0",
+        version: "0.3.4",
      });
      dependencies.push({
        name: "llama-index-embeddings-gemini",
-        version: ">=0.3.0,<0.4.0",
+        version: "^0.2.0",
      });
      break;
    case "mistral":
      dependencies.push({
        name: "llama-index-llms-mistralai",
-        version: ">=0.4.0,<0.5.0",
+        version: "0.2.1",
      });
      dependencies.push({
        name: "llama-index-embeddings-mistralai",
-        version: ">=0.3.0,<0.4.0",
+        version: "0.2.0",
      });
      break;
    case "azure-openai":
      dependencies.push({
        name: "llama-index-llms-azure-openai",
-        version: ">=0.3.0,<0.4.0",
+        version: "0.2.0",
      });
      dependencies.push({
        name: "llama-index-embeddings-azure-openai",
-        version: ">=0.3.0,<0.4.0",
+        version: "0.2.4",
      });
      break;
    case "huggingface":
      dependencies.push({
        name: "llama-index-llms-huggingface",
-        version: ">=0.5.0,<0.6.0",
+        version: "^0.3.5",
      });
      dependencies.push({
        name: "llama-index-embeddings-huggingface",
-        version: ">=0.5.0,<0.6.0",
+        version: "^0.3.1",
      });
      dependencies.push({
        name: "optimum",
-        version: ">=1.23.3,<2.0.0",
+        version: "^1.23.3",
        extras: ["onnxruntime"],
      });
      break;
    case "t-systems":
      dependencies.push({
        name: "llama-index-agent-openai",
-        version: ">=0.4.0,<0.5.0",
+        version: "0.3.0",
      });
      dependencies.push({
        name: "llama-index-llms-openai-like",
-        version: ">=0.3.0,<0.4.0",
+        version: "0.2.0",
      });
      break;
  }

-  if (observability && observability !== "none") {
-    if (observability === "traceloop") {
-      dependencies.push({
-        name: "traceloop-sdk",
-        version: ">=0.15.11",
-      });
+  return dependencies;
+};
+
+const mergePoetryDependencies = (
+  dependencies: Dependency[],
+  existingDependencies: Record<string, Omit<Dependency, "name"> | string>,
+) => {
+  for (const dependency of dependencies) {
+    let value = existingDependencies[dependency.name] ?? {};
+
+    // default string value is equal to attribute "version"
+    if (typeof value === "string") {
+      value = { version: value };
    }
-    if (observability === "llamatrace") {
-      dependencies.push({
-        name: "llama-index-callbacks-arize-phoenix",
-        version: ">=0.3.0,<0.4.0",
-      });
+
+    value.version = dependency.version ?? value.version;
+    value.extras = dependency.extras ?? value.extras;
+
+    // Merge constraints if they exist
+    if (dependency.constraints) {
+      value = { ...value, ...dependency.constraints };
+    }
+
+    if (value.version === undefined) {
+      throw new Error(
+        `Dependency "${dependency.name}" is missing attribute "version"!`,
+      );
+    }
+
+    // Serialize as object if there are any additional properties
+    if (Object.keys(value).length > 1) {
+      existingDependencies[dependency.name] = value;
+    } else {
+      // Otherwise, serialize just the version string
+      existingDependencies[dependency.name] = value.version;
    }
  }
-
-  return dependencies;
 };

 const copyRouterCode = async (root: string, tools: Tool[]) => {
@@ -303,100 +329,19 @@ export const addDependencies = async (
    // Parse toml file
    const file = path.join(projectDir, FILENAME);
    const fileContent = await fs.readFile(file, "utf8");
-    let fileParsed: any;
-    try {
-      fileParsed = parse(fileContent);
-    } catch (parseError) {
-      console.error(`Error parsing ${FILENAME}:`, parseError);
-      throw new Error(
-        `Failed to parse ${FILENAME}. Please ensure it's valid TOML.`,
-      );
-    }
+    const fileParsed = parse(fileContent);

-    // Ensure [project] and [project.dependencies] exist
-    if (!fileParsed.project) {
-      fileParsed.project = {};
-    }
-    if (
-      !fileParsed.project.dependencies ||
-      !Array.isArray(fileParsed.project.dependencies)
-    ) {
-      // If dependencies exist but aren't an array, log a warning or error.
-      // For now, we'll overwrite it, assuming the intent is to use the standard array format.
-      console.warn(
-        `[project.dependencies] in ${FILENAME} is not an array. It will be overwritten.`,
-      );
-      fileParsed.project.dependencies = [];
-    }
-
-    const existingDependencies: string[] = fileParsed.project.dependencies;
-    const addedDeps: string[] = [];
-    const updatedDeps: string[] = [];
-
-    // Add or update dependencies
-    for (const newDep of dependencies) {
-      let depString = newDep.name;
-      if (newDep.extras && newDep.extras.length > 0) {
-        depString += `[${newDep.extras.join(",")}]`;
-      }
-      if (newDep.version) {
-        depString += newDep.version;
-      }
-
-      let found = false;
-      for (let i = 0; i < existingDependencies.length; i++) {
-        const existingDepNameMatch =
-          existingDependencies[i].match(/^([a-zA-Z0-9._-]+)/);
-        if (
-          existingDepNameMatch &&
-          existingDepNameMatch[1].toLowerCase() === depString.toLowerCase()
-        ) {
-          // Found existing dependency, update it
-          if (existingDependencies[i] !== depString) {
-            updatedDeps.push(`${existingDependencies[i]} -> ${depString}`);
-            existingDependencies[i] = depString;
-          }
-          found = true;
-          break;
-        }
-      }
-
-      if (!found) {
-        // Add new dependency
-        existingDependencies.push(depString);
-        addedDeps.push(depString);
-      }
-      // Handle python version constraints separately (if any)
-      if (newDep.constraints?.python) {
-        if (
-          !fileParsed.project["requires-python"] ||
-          fileParsed.project["requires-python"] !== newDep.constraints.python
-        ) {
-          // This simple overwrite might not be ideal; merging constraints is complex.
-          // For now, let's just set it if the new dependency has one.
-          console.log(
-            `Setting requires-python = "${newDep.constraints.python}" from dependency ${newDep.name}`,
-          );
-          fileParsed.project["requires-python"] = newDep.constraints.python;
-        }
-      }
-    }
+    // Modify toml dependencies
+    const tool = fileParsed.tool as any;
+    const existingDependencies = tool.poetry.dependencies;
+    mergePoetryDependencies(dependencies, existingDependencies);

    // Write toml file
    const newFileContent = stringify(fileParsed);
    await fs.writeFile(file, newFileContent);

-    if (addedDeps.length > 0) {
-      console.log(`\nAdded dependencies to ${cyan(FILENAME)}:`);
-      addedDeps.forEach((dep) => console.log(`  ${dep}`));
-    }
-    if (updatedDeps.length > 0) {
-      console.log(`\nUpdated dependencies in ${cyan(FILENAME)}:`);
-      updatedDeps.forEach((dep) => console.log(`  ${dep}`));
-    }
-    if (addedDeps.length > 0 || updatedDeps.length > 0) {
-      console.log(""); // Newline for spacing
-    }
+    const dependenciesString = dependencies.map((d) => d.name).join(", ");
+    console.log(`\nAdded ${dependenciesString} to ${cyan(FILENAME)}\n`);
  } catch (error) {
    console.log(
      `Error while updating dependencies for Poetry project file ${FILENAME}\n`,
@@ -405,16 +350,18 @@ export const addDependencies = async (
  }
 };

-export const installPythonDependencies = () => {
-  if (isUvAvailable()) {
+export const installPythonDependencies = (
+  { noRoot }: { noRoot: boolean } = { noRoot: false },
+) => {
+  if (isPoetryAvailable()) {
    console.log(
-      `Installing Python dependencies using uv. This may take a while...`,
+      `Installing python dependencies using poetry. This may take a while...`,
    );
-    const installSuccessful = tryUvSync();
+    const installSuccessful = tryPoetryInstall(noRoot);
    if (!installSuccessful) {
      console.error(
        red(
-          "Installing dependencies using uv failed. Please check the error log above and ensure uv is installed correctly.",
+          "Installing dependencies using poetry failed. Please check error log above and try running create-llama again.",
        ),
      );
      process.exit(1);
@@ -422,34 +369,57 @@ export const installPythonDependencies = () => {
  } else {
    console.error(
      red(
-        `uv is not available in the current environment. Please check ${terminalLink(
-          "uv Installation",
-          `https://github.com/astral-sh/uv#installation`,
-        )} to install uv first, then run create-llama again.`,
+        `Poetry is not available in the current environment. Please check ${terminalLink(
+          "Poetry Installation",
+          `https://python-poetry.org/docs/#installation`,
+        )} to install poetry first, then run create-llama again.`,
      ),
    );
    process.exit(1);
  }
 };

-const installLegacyPythonTemplate = async ({
+export const installPythonTemplate = async ({
+  appName,
  root,
  template,
+  framework,
  vectorDb,
+  postInstallAction,
+  modelConfig,
  dataSources,
  tools,
+  useLlamaParse,
  useCase,
  observability,
 }: Pick<
  InstallTemplateArgs,
+  | "appName"
  | "root"
  | "template"
+  | "framework"
  | "vectorDb"
+  | "postInstallAction"
+  | "modelConfig"
  | "dataSources"
  | "tools"
+  | "useLlamaParse"
  | "useCase"
  | "observability"
 >) => {
+  console.log("\nInitializing Python project with template:", template, "\n");
+  let templatePath;
+  if (template === "reflex") {
+    templatePath = path.join(templatesDir, "types", "reflex");
+  } else {
+    templatePath = path.join(templatesDir, "types", "streaming", framework);
+  }
+  await copy("**", root, {
+    parents: true,
+    cwd: templatePath,
+    rename: assetRelocator,
+  });
+
  const compPath = path.join(templatesDir, "components");
  const enginePath = path.join(root, "app", "engine");

@@ -539,7 +509,34 @@ const installLegacyPythonTemplate = async ({
    }
  }

+  console.log("Adding additional dependencies");
+
+  const addOnDependencies = getAdditionalDependencies(
+    modelConfig,
+    vectorDb,
+    dataSources,
+    tools,
+    template,
+  );
+
  if (observability && observability !== "none") {
+    if (observability === "traceloop") {
+      addOnDependencies.push({
+        name: "traceloop-sdk",
+        version: "^0.15.11",
+      });
+    }
+
+    if (observability === "llamatrace") {
+      addOnDependencies.push({
+        name: "llama-index-callbacks-arize-phoenix",
+        version: "^0.3.0",
+        constraints: {
+          python: ">=3.11,<3.13",
+        },
+      });
+    }
+
    const templateObservabilityPath = path.join(
      templatesDir,
      "components",
@@ -551,134 +548,6 @@ const installLegacyPythonTemplate = async ({
      cwd: templateObservabilityPath,
    });
  }
-};
-
-const installLlamaIndexServerTemplate = async ({
-  root,
-  useCase,
-  useLlamaParse,
-}: Pick<InstallTemplateArgs, "root" | "useCase" | "useLlamaParse">) => {
-  if (!useCase) {
-    console.log(
-      red(
-        `There is no use case selected. Please pick a use case to use via --use-case flag.`,
-      ),
-    );
-    process.exit(1);
-  }
-
-  await copy("*.py", path.join(root, "app"), {
-    parents: true,
-    cwd: path.join(templatesDir, "components", "use-cases", "python", useCase),
-  });
-
-  // Copy custom UI component code
-  await copy(`*`, path.join(root, "components"), {
-    parents: true,
-    cwd: path.join(templatesDir, "components", "ui", "use-cases", useCase),
-  });
-
-  if (useLlamaParse) {
-    await copy("index.py", path.join(root, "app"), {
-      parents: true,
-      cwd: path.join(
-        templatesDir,
-        "components",
-        "vectordbs",
-        "llamaindexserver",
-        "llamacloud",
-        "python",
-      ),
-    });
-    // TODO: Consider moving generate.py to app folder.
-    await copy("generate.py", path.join(root), {
-      parents: true,
-      cwd: path.join(
-        templatesDir,
-        "components",
-        "vectordbs",
-        "llamaindexserver",
-        "llamacloud",
-        "python",
-      ),
-    });
-  }
-  // Copy README.md
-  await copy("README-template.md", path.join(root), {
-    parents: true,
-    cwd: path.join(templatesDir, "components", "use-cases", "python", useCase),
-    rename: assetRelocator,
-  });
-};
-
-export const installPythonTemplate = async ({
-  appName,
-  root,
-  template,
-  framework,
-  vectorDb,
-  postInstallAction,
-  modelConfig,
-  dataSources,
-  tools,
-  useLlamaParse,
-  useCase,
-  observability,
-}: Pick<
-  InstallTemplateArgs,
-  | "appName"
-  | "root"
-  | "template"
-  | "framework"
-  | "vectorDb"
-  | "postInstallAction"
-  | "modelConfig"
-  | "dataSources"
-  | "tools"
-  | "useLlamaParse"
-  | "useCase"
-  | "observability"
->) => {
-  console.log("\nInitializing Python project with template:", template, "\n");
-  let templatePath;
-  if (template === "reflex") {
-    templatePath = path.join(templatesDir, "types", "reflex");
-  } else {
-    templatePath = path.join(templatesDir, "types", template, framework);
-  }
-  await copy("**", root, {
-    parents: true,
-    cwd: templatePath,
-    rename: assetRelocator,
-  });
-
-  if (template === "llamaindexserver") {
-    await installLlamaIndexServerTemplate({
-      root,
-      useCase,
-      useLlamaParse,
-    });
-  } else {
-    await installLegacyPythonTemplate({
-      root,
-      template,
-      vectorDb,
-      dataSources,
-      tools,
-      useCase,
-      observability,
-    });
-  }
-
-  console.log("Adding additional dependencies");
-  const addOnDependencies = getAdditionalDependencies(
-    modelConfig,
-    vectorDb,
-    dataSources,
-    tools,
-    template,
-    observability,
-  );

  await addDependencies(root, addOnDependencies);

@@ -34,24 +34,14 @@ export function runReflexApp(appPath: string, port: number) {
    "--frontend-port",
    port.toString(),
  ];
-  return createProcess("uv", commandArgs, {
+  return createProcess("poetry", commandArgs, {
    stdio: "inherit",
    cwd: appPath,
  });
 }

-export function runFastAPIApp(
-  appPath: string,
-  port: number,
-  template: TemplateType,
-) {
-  let commandArgs: string[];
-  if (template === "streaming") {
-    commandArgs = ["run", "dev"];
-  } else {
-    commandArgs = ["run", "fastapi", "dev", "--port", `${port}`];
-  }
-  return createProcess("uv", commandArgs, {
+export function runFastAPIApp(appPath: string, port: number) {
+  return createProcess("poetry", ["run", "dev"], {
    stdio: "inherit",
    cwd: appPath,
    env: { ...process.env, APP_PORT: `${port}` },
@@ -83,7 +73,7 @@ export async function runApp(
        : framework === "fastapi"
          ? runFastAPIApp
          : runTSApp;
-    await appRunner(appPath, port || defaultPort, template);
+    await appRunner(appPath, port || defaultPort);
  } catch (error) {
    console.error("Failed to run app:", error);
    throw error;
@@ -41,7 +41,7 @@ export const supportedTools: Tool[] = [
    dependencies: [
      {
        name: "llama-index-tools-google",
-        version: ">=0.3.0,<0.4.0",
+        version: "^0.3.0",
      },
    ],
    supportedFrameworks: ["fastapi"],
@@ -62,7 +62,7 @@ export const supportedTools: Tool[] = [
    dependencies: [
      {
        name: "duckduckgo-search",
-        version: ">=6.3.5,<7.0.0",
+        version: "^6.3.5",
      },
    ],
    supportedFrameworks: ["fastapi"], // TODO: Re-enable this tool once the duck-duck-scrape TypeScript library works again
@@ -82,7 +82,7 @@ For better results, you can specify the region parameter to get results from a s
    dependencies: [
      {
        name: "llama-index-tools-wikipedia",
-        version: ">=0.3.0,<0.4.0",
+        version: "^0.3.0",
      },
    ],
    supportedFrameworks: ["fastapi", "express", "nextjs"],
@@ -102,11 +102,11 @@ For better results, you can specify the region parameter to get results from a s
    dependencies: [
      {
        name: "xhtml2pdf",
-        version: ">=0.2.14,<0.3.0",
+        version: "^0.2.14",
      },
      {
        name: "markdown",
-        version: ">=3.7.0,<4.0.0",
+        version: "^3.7",
      },
    ],
    type: ToolType.LOCAL,
@@ -124,7 +124,7 @@ For better results, you can specify the region parameter to get results from a s
    dependencies: [
      {
        name: "e2b_code_interpreter",
-        version: ">=1.1.1,<1.2.0",
+        version: "1.0.3",
      },
    ],
    supportedFrameworks: ["fastapi", "express", "nextjs"],
@@ -155,7 +155,7 @@ For better results, you can specify the region parameter to get results from a s
    dependencies: [
      {
        name: "e2b_code_interpreter",
-        version: ">=1.1.1,<1.2.0",
+        version: "1.0.3",
      },
    ],
    supportedFrameworks: ["fastapi", "express", "nextjs"],
@@ -184,7 +184,7 @@ For better results, you can specify the region parameter to get results from a s
      },
      {
        name: "jsonschema",
-        version: ">=4.22.0,<5.0.0",
+        version: "^4.22.0",
      },
      {
        name: "llama-index-tools-requests",
@@ -247,11 +247,11 @@ For better results, you can specify the region parameter to get results from a s
    dependencies: [
      {
        name: "pandas",
-        version: ">=2.2.3,<3.0.0",
+        version: "^2.2.3",
      },
      {
        name: "tabulate",
-        version: ">=0.9.0,<1.0.0",
+        version: "^0.9.0",
      },
    ],
  },
@@ -24,8 +24,7 @@ export type TemplateType =
  | "community"
  | "llamapack"
  | "multiagent"
-  | "reflex"
-  | "llamaindexserver";
+  | "reflex";
 export type TemplateFramework = "nextjs" | "express" | "fastapi";
 export type TemplateUI = "html" | "shadcn";
 export type TemplateVectorDB =
@@ -56,10 +55,7 @@ export type TemplateUseCase =
  | "deep_research"
  | "form_filling"
  | "extractor"
-  | "contract_review"
-  | "agentic_rag"
-  | "code_generator"
-  | "document_generator";
+  | "contract_review";
 // Config for both file and folder
 export type FileSourceConfig =
  | {
@@ -8,101 +8,42 @@ import { templatesDir } from "./dir";
 import { PackageManager } from "./get-pkg-manager";
 import { InstallTemplateArgs, ModelProvider, TemplateVectorDB } from "./types";

-const installLlamaIndexServerTemplate = async ({
-  root,
-  useCase,
-  vectorDb,
-}: Pick<InstallTemplateArgs, "root" | "useCase" | "vectorDb">) => {
-  if (!useCase) {
-    console.log(
-      red(
-        `There is no use case selected. Please pick a use case to use via --use-case flag.`,
-      ),
-    );
-    process.exit(1);
-  }
-
-  if (!vectorDb) {
-    console.log(
-      red(
-        `There is no vector db selected. Please pick a vector db to use via --vector-db flag.`,
-      ),
-    );
-    process.exit(1);
-  }
-
-  await copy("**", path.join(root), {
-    cwd: path.join(
-      templatesDir,
-      "components",
-      "use-cases",
-      "typescript",
-      useCase,
-    ),
-    rename: assetRelocator,
-  });
-
-  // copy workflow UI components to output/components folder
-  await copy("*", path.join(root, "components"), {
-    parents: true,
-    cwd: path.join(templatesDir, "components", "ui", "use-cases", useCase),
-  });
-
-  // Override generate.ts if workflow use case doesn't use custom UI
-  if (vectorDb === "llamacloud") {
-    await copy("generate.ts", path.join(root, "src"), {
-      parents: true,
-      cwd: path.join(
-        templatesDir,
-        "components",
-        "vectordbs",
-        "llamaindexserver",
-        "llamacloud",
-        "typescript",
-      ),
-    });
-
-    await copy("index.ts", path.join(root, "src", "app"), {
-      parents: true,
-      cwd: path.join(
-        templatesDir,
-        "components",
-        "vectordbs",
-        "llamaindexserver",
-        "llamacloud",
-        "typescript",
-      ),
-      rename: () => "data.ts",
-    });
-  }
-
-  // Simplify use case code
-  if (useCase === "code_generator" || useCase === "document_generator") {
-    // Artifact use case doesn't use index.
-    // We don't need data.ts, generate.ts
-    await fs.rm(path.join(root, "src", "app", "data.ts"));
-    // TODO: Remove generate index in generate.ts and package.json if possible
-  }
-};
-
-const installLegacyTSTemplate = async ({
+/**
+ * Install a LlamaIndex internal template to a given `root` directory.
+ */
+export const installTSTemplate = async ({
+  appName,
  root,
+  packageManager,
+  isOnline,
  template,
-  backend,
  framework,
  ui,
  vectorDb,
+  postInstallAction,
+  backend,
  observability,
  tools,
  dataSources,
  useLlamaParse,
  useCase,
  modelConfig,
-  relativeEngineDestPath,
-}: InstallTemplateArgs & {
-  backend: boolean;
-  relativeEngineDestPath: string;
-}) => {
+}: InstallTemplateArgs & { backend: boolean }) => {
+  console.log(bold(`Using ${packageManager}.`));
+
+  /**
+   * Copy the template files to the target directory.
+   */
+  console.log("\nInitializing project with template:", template, "\n");
+  const templatePath = path.join(templatesDir, "types", "streaming", framework);
+  const copySource = ["**"];
+
+  await copy(copySource, root, {
+    parents: true,
+    cwd: templatePath,
+    rename: assetRelocator,
+  });
+
  /**
   * If next.js is used, update its configuration if necessary
   */
@@ -157,6 +98,10 @@ const installLegacyTSTemplate = async ({
  }

  const compPath = path.join(templatesDir, "components");
+  const relativeEngineDestPath =
+    framework === "nextjs"
+      ? path.join("app", "api", "chat")
+      : path.join("src", "controllers");
  const enginePath = path.join(root, relativeEngineDestPath, "engine");

  // copy llamaindex code for TS templates
@@ -291,75 +236,6 @@ const installLegacyTSTemplate = async ({
    await fs.rm(path.join(root, "app", "api"), { recursive: true });
    await fs.rm(path.join(root, "config"), { recursive: true, force: true });
  }
-};
-
-/**
- * Install a LlamaIndex internal template to a given `root` directory.
- */
-export const installTSTemplate = async ({
-  appName,
-  root,
-  packageManager,
-  isOnline,
-  template,
-  framework,
-  ui,
-  vectorDb,
-  postInstallAction,
-  backend,
-  observability,
-  tools,
-  dataSources,
-  useLlamaParse,
-  useCase,
-  modelConfig,
-}: InstallTemplateArgs & { backend: boolean }) => {
-  console.log(bold(`Using ${packageManager}.`));
-
-  /**
-   * Copy the template files to the target directory.
-   */
-  console.log("\nInitializing project with template:", template, "\n");
-  const templatePath = path.join(templatesDir, "types", template, framework);
-  const copySource = ["**"];
-
-  await copy(copySource, root, {
-    parents: true,
-    cwd: templatePath,
-    rename: assetRelocator,
-  });
-
-  const relativeEngineDestPath =
-    framework === "nextjs"
-      ? path.join("app", "api", "chat")
-      : path.join("src", "controllers");
-
-  if (template === "llamaindexserver") {
-    await installLlamaIndexServerTemplate({
-      root,
-      useCase,
-      vectorDb,
-    });
-  } else {
-    await installLegacyTSTemplate({
-      appName,
-      root,
-      packageManager,
-      isOnline,
-      template,
-      backend,
-      framework,
-      ui,
-      vectorDb,
-      observability,
-      tools,
-      dataSources,
-      useLlamaParse,
-      useCase,
-      modelConfig,
-      relativeEngineDestPath,
-    });
-  }

  const packageJson = await updatePackageJson({
    root,
@@ -372,7 +248,6 @@ export const installTSTemplate = async ({
    vectorDb,
    backend,
    modelConfig,
-    template,
  });

  if (
@@ -387,27 +262,27 @@ const providerDependencies: {
  [key in ModelProvider]?: Record<string, string>;
 } = {
  openai: {
-    "@llamaindex/openai": "^0.3.7",
+    "@llamaindex/openai": "^0.1.52",
  },
  gemini: {
-    "@llamaindex/google": "^0.2.0",
+    "@llamaindex/google": "^0.0.7",
  },
  ollama: {
-    "@llamaindex/ollama": "^0.1.0",
+    "@llamaindex/ollama": "^0.0.40",
  },
  mistral: {
-    "@llamaindex/mistral": "^0.2.0",
+    "@llamaindex/mistral": "^0.0.5",
  },
  "azure-openai": {
-    "@llamaindex/openai": "^0.2.0",
+    "@llamaindex/openai": "^0.1.52",
  },
  groq: {
-    "@llamaindex/groq": "^0.0.61",
-    "@llamaindex/huggingface": "^0.1.0", // groq uses huggingface as default embedding model
+    "@llamaindex/groq": "^0.0.51",
+    "@llamaindex/huggingface": "^0.0.36", // groq uses huggingface as default embedding model
  },
  anthropic: {
-    "@llamaindex/anthropic": "^0.3.0",
-    "@llamaindex/huggingface": "^0.1.0", // anthropic uses huggingface as default embedding model
+    "@llamaindex/anthropic": "^0.1.0",
+    "@llamaindex/huggingface": "^0.0.36", // anthropic uses huggingface as default embedding model
  },
 };

@@ -456,7 +331,6 @@ async function updatePackageJson({
  vectorDb,
  backend,
  modelConfig,
-  template,
 }: Pick<
  InstallTemplateArgs,
  | "root"
@@ -467,7 +341,6 @@ async function updatePackageJson({
  | "observability"
  | "vectorDb"
  | "modelConfig"
-  | "template"
 > & {
  relativeEngineDestPath: string;
  backend: boolean;
@@ -479,7 +352,7 @@ async function updatePackageJson({
  packageJson.name = appName;
  packageJson.version = "0.1.0";

-  if (relativeEngineDestPath && template !== "llamaindexserver") {
+  if (relativeEngineDestPath) {
    // TODO: move script to {root}/scripts for all frameworks
    // add generate script if using context engine
    packageJson.scripts = {
@@ -513,7 +386,7 @@ async function updatePackageJson({
  if (backend) {
    packageJson.dependencies = {
      ...packageJson.dependencies,
-      "@llamaindex/readers": "^3.1.3",
+      "@llamaindex/readers": "^2.0.0",
    };

    if (vectorDb && vectorDb in vectorDbDependencies) {
@@ -1,3 +1,4 @@
+// eslint-disable-next-line import/no-extraneous-dependencies
 import validateProjectName from "validate-npm-package-name";

 export function validateNpmName(name: string): {
@@ -1,3 +1,4 @@
+/* eslint-disable import/no-extraneous-dependencies */
 import { execSync } from "child_process";
 import { Command } from "commander";
 import fs from "fs";
@@ -196,7 +197,7 @@ const program = new Command(packageJson.name)
    "--pro",
    `

-  Deprecated: Allow interactive selection of all features.
+  Allow interactive selection of all features.
 `,
    false,
  )
@@ -6,7 +6,7 @@ LlamaIndexServer is a FastAPI-based application that allows you to quickly launc

 - Serving a workflow as a chatbot
 - Built on FastAPI for high performance and easy API development
- Optional built-in chat UI with extendable UI components
+- Optional built-in chat UI
 - Prebuilt development code

 ## Installation
@@ -43,10 +43,8 @@ def create_workflow() -> Workflow:
 app = LlamaIndexServer(
    workflow_factory=create_workflow,  # Supports Workflow or AgentWorkflow
    env="dev",  # Enable development mode
-    ui_config={ # Configure the chat UI, optional
-        "app_title": "Weather Bot",
-        "starter_questions": ["What is the weather in LA?", "Will it rain in SF?"],
-    },
+    include_ui=True,  # Include chat UI
+    starter_questions=["What can you do?", "How do I use this?"],
    verbose=True
 )
 ```
@@ -72,46 +70,16 @@ app = LlamaIndexServer(

 The LlamaIndexServer accepts the following configuration parameters:

- `workflow_factory`: A callable that creates a workflow instance for each request. See [Workflow factory contract](#workflow-factory-contract) for more details.
+- `workflow_factory`: A callable that creates a workflow instance for each request
 - `logger`: Optional logger instance (defaults to uvicorn logger)
 - `use_default_routers`: Whether to include default routers (chat, static file serving)
 - `env`: Environment setting ('dev' enables CORS and UI by default)
- `ui_config`: UI configuration as a dictionary or UIConfig object with options:
-  - `enabled`: Whether to enable the chat UI (default: True)
-  - `app_title`: The title of the chat application (default: "LlamaIndex Server")
-  - `starter_questions`: List of starter questions for the chat UI (default: None)
-  - `ui_path`: Path for downloaded UI static files (default: ".ui")
-  - `component_dir`: The directory for custom UI components rendering events emitted by the workflow. The default is None, which does not render custom UI components.
-  - `llamacloud_index_selector`: Whether to show the LlamaCloud index selector in the chat UI (default: False). Requires `LLAMA_CLOUD_API_KEY` to be set.
-  - `dev_mode`: When enabled, you can update workflow code in the UI and see the changes immediately. It's currently in beta and only supports updating workflow code at `app/workflow.py`. You might also need to set `env="dev"` and start the server with the reload feature enabled.
+- `include_ui`: Whether to include the chat UI
+- `starter_questions`: List of starter questions for the chat UI
 - `verbose`: Enable verbose logging
 - `api_prefix`: API route prefix (default: "/api")
 - `server_url`: The deployment URL of the server (default is None)
-
-## Workflow factory contract
-
-The `workflow_factory` provided will be called for each chat request to initialize a new workflow instance. Additionally, we provide the [ChatRequest](https://github.com/run-llama/create-llama/blob/afe9e9fc16427d20e1dfb635a45e7ed4b46285cb/python/llama-index-server/llama_index/server/api/models.py#L32) object, which includes the request information that is helpful for initializing the workflow. For example:
-```python
-def create_workflow(chat_request: ChatRequest) -> Workflow:
-    # using messages from the chat request to initialize the workflow
-    return MyCustomWorkflow(chat_request.messages)
-```
-
-Your workflow will be executed once for each chat request with the following input parameters are included in workflow's `StartEvent`:
- `user_msg` [str]: The current user message
- `chat_history` [list[[ChatMessage](https://docs.llamaindex.ai/en/stable/api_reference/prompts/#llama_index.core.prompts.ChatMessage)]]: All the previous messages of the conversation
-
-Example:
-```python
-@step
-def handle_start_event(ev: StartEvent) -> MyNextEvent:
-    user_msg = ev.user_msg
-    chat_history = ev.chat_history
-    ...
-```
-
-Your workflows can emit `UIEvent` events to render [Custom UI Components](https://github.com/run-llama/create-llama/blob/main/python/llama-index-server/docs/custom_ui_component.md) in the chat UI to improve the user experience.
-Furthermore, you can send `ArtifactEvent` events to render code or document [Artifacts](https://github.com/run-llama/create-llama/blob/main/python/llama-index-server/docs/custom_artifact_event.md) in a dedicated Canvas panel in the chat UI.
+- `ui_path`: Path for downloaded UI static files (default: ".ui")

 ## Default Routers and Features

@@ -141,20 +109,6 @@ In development mode (`env="dev"`), the server:
 - Automatically includes the chat UI
 - Provides more verbose logging

-### Workflow Editor (Beta)
-
-In development mode, you can set `dev_mode` to `True` in the UI configuration to enable the workflow editor, which allows you to edit the workflow code directly in the browser.
-
-```python
-app = LlamaIndexServer(
-    workflow_factory=create_workflow,
-    env="dev",
-    ui_config={"dev_mode": True},
-)
-```
-
-**Note**: The workflow editor is currently in beta and only supports updating LlamaIndexServer projects created with [create-llama](https://github.com/run-llama/create-llama/). You also need to start the server via `fastapi dev` so that the server can hot reload the workflow code.
-
 ## API Endpoints

 The server provides the following default endpoints:
@@ -165,10 +119,11 @@ The server provides the following default endpoints:

 ## Best Practices

-1. Use environment variables for sensitive configuration
-2. Enable verbose logging during development
-3. Configure CORS appropriately for your deployment environment
-4. Use starter questions to guide users in the chat UI
+1. Always provide a workflow factory that creates fresh workflow instances
+2. Use environment variables for sensitive configuration
+3. Enable verbose logging during development
+4. Configure CORS appropriately for your deployment environment
+5. Use starter questions to guide users in the chat UI

 ## Getting Started with a New Project

@@ -0,0 +1,3 @@
+from .server import LlamaIndexServer
+
+__all__ = ["LlamaIndexServer"]
@@ -1,14 +1,13 @@
 import logging
 import os
 from enum import Enum
-from typing import Any, Dict, List, Literal, Optional, Union
-
-from pydantic import BaseModel, Field, field_validator
+from typing import Any, Dict, List, Optional

 from llama_index.core.schema import NodeWithScore
 from llama_index.core.types import ChatMessage, MessageRole
 from llama_index.core.workflow import Event
 from llama_index.server.settings import server_settings
+from pydantic import BaseModel, Field, field_validator

 logger = logging.getLogger("uvicorn")

@@ -23,7 +22,6 @@ class ChatConfig(BaseModel):
 class ChatAPIMessage(BaseModel):
    role: MessageRole
    content: str
-    annotations: Optional[List[Any]] = None

    def to_llamaindex_message(self) -> ChatMessage:
        return ChatMessage(role=self.role, content=self.content)
@@ -136,71 +134,3 @@ class SourceNodes(BaseModel):
        cls, source_nodes: List[NodeWithScore]
    ) -> List["SourceNodes"]:
        return [cls.from_source_node(node) for node in source_nodes]
-
-
-class ComponentDefinition(BaseModel):
-    type: str
-    code: str
-    filename: str
-
-
-class UIEvent(Event):
-    type: str
-    data: BaseModel
-
-    def to_response(self) -> dict:
-        return {
-            "type": self.type,
-            "data": self.data.model_dump(),
-        }
-
-
-class ArtifactType(str, Enum):
-    CODE = "code"
-    DOCUMENT = "document"
-
-
-class CodeArtifactData(BaseModel):
-    file_name: str
-    code: str
-    language: str
-
-
-class DocumentArtifactData(BaseModel):
-    title: str
-    content: str
-    type: Literal["markdown", "html"]
-
-
-class Artifact(BaseModel):
-    created_at: Optional[int] = None
-    type: ArtifactType
-    data: Union[CodeArtifactData, DocumentArtifactData]
-
-    @classmethod
-    def from_message(cls, message: ChatAPIMessage) -> Optional["Artifact"]:
-        if not message.annotations or not isinstance(message.annotations, list):
-            return None
-
-        for annotation in message.annotations:
-            if isinstance(annotation, dict) and annotation.get("type") == "artifact":
-                try:
-                    artifact = cls.model_validate(annotation.get("data"))
-                    return artifact
-                except Exception as e:
-                    logger.warning(
-                        f"Failed to parse artifact from annotation: {annotation}. Error: {e}"
-                    )
-
-        return None
-
-
-class ArtifactEvent(Event):
-    type: str = "artifact"
-    data: Artifact
-
-    def to_response(self) -> dict:
-        return {
-            "type": self.type,
-            "data": self.data.model_dump(),
-        }
@@ -6,19 +6,14 @@ from typing import AsyncGenerator, Callable, Union

 from fastapi import APIRouter, BackgroundTasks, HTTPException
 from fastapi.responses import StreamingResponse
-
-from llama_index.core.agent.workflow.workflow_events import (
-    AgentInput,
-    AgentSetup,
-    AgentStream,
-)
+from llama_index.core.agent.workflow.workflow_events import AgentStream
 from llama_index.core.workflow import StopEvent, Workflow
 from llama_index.server.api.callbacks import (
-    EventCallback,
-    LlamaCloudFileDownload,
    SourceNodesFromToolCall,
    SuggestNextQuestions,
 )
+from llama_index.server.api.callbacks.base import EventCallback
+from llama_index.server.api.callbacks.llamacloud import LlamaCloudFileDownload
 from llama_index.server.api.callbacks.stream_handler import StreamHandler
 from llama_index.server.api.models import ChatRequest
 from llama_index.server.api.utils.vercel_stream import VercelStreamResponse
@@ -104,10 +99,7 @@ async def _stream_content(
        event: Union[AgentStream, StopEvent],
    ) -> AsyncGenerator[str, None]:
        if isinstance(event, AgentStream):
-            # Normally, if the stream is a tool call, the delta is always empty
-            # so it's not a text stream.
-            if len(event.tool_calls) == 0:
-                yield event.delta
+            yield event.delta
        elif isinstance(event, StopEvent):
            if isinstance(event.result, str):
                yield event.result
@@ -118,8 +110,15 @@ async def _stream_content(
                    elif hasattr(chunk, "delta") and chunk.delta:
                        yield chunk.delta

+    stream_started = False
    try:
        async for event in handler.stream_events():
+            if not stream_started:
+                # Start the stream with an empty message
+                stream_started = True
+                yield VercelStreamResponse.convert_text("")
+
+            # Handle different types of events
            if isinstance(event, (AgentStream, StopEvent)):
                async for chunk in _text_stream(event):
                    handler.accumulate_text(chunk)
@@ -130,14 +129,12 @@ async def _stream_content(
                event_response = event.to_response()
                yield VercelStreamResponse.convert_data(event_response)
            else:
-                # Ignore unnecessary agent workflow events
-                if not isinstance(event, (AgentInput, AgentSetup)):
-                    yield VercelStreamResponse.convert_data(event.model_dump())
+                yield VercelStreamResponse.convert_data(event.model_dump())

    except asyncio.CancelledError:
        logger.warning("Client cancelled the request!")
        await handler.cancel_run()
    except Exception as e:
-        logger.error(f"Error in stream response: {e}", exc_info=True)
+        logger.error(f"Error in stream response: {e}")
        yield VercelStreamResponse.convert_error(str(e))
        await handler.cancel_run()
@@ -5,7 +5,7 @@ from typing import Optional

 import requests

-CHAT_UI_VERSION = "0.2.1"
+CHAT_UI_VERSION = "0.0.6"


 def download_chat_ui(
@@ -0,0 +1,184 @@
+import json
+import logging
+import os
+from typing import Any, Callable, Optional
+
+from fastapi import FastAPI
+from fastapi.middleware.cors import CORSMiddleware
+from fastapi.staticfiles import StaticFiles
+from llama_index.core.workflow import Workflow
+from llama_index.server.api.routers.chat import chat_router
+from llama_index.server.chat_ui import download_chat_ui
+from llama_index.server.settings import server_settings
+
+
+class LlamaIndexServer(FastAPI):
+    workflow_factory: Callable[..., Workflow]
+    include_ui: Optional[bool]
+    starter_questions: Optional[list[str]]
+    verbose: bool = False
+    ui_path: str = ".ui"
+
+    def __init__(
+        self,
+        workflow_factory: Callable[..., Workflow],
+        logger: Optional[logging.Logger] = None,
+        use_default_routers: Optional[bool] = True,
+        env: Optional[str] = None,
+        include_ui: Optional[bool] = None,
+        starter_questions: Optional[list[str]] = None,
+        server_url: Optional[str] = None,
+        api_prefix: Optional[str] = None,
+        verbose: bool = False,
+        *args: Any,
+        **kwargs: Any,
+    ):
+        """
+        Initialize the LlamaIndexServer.
+
+        Args:
+            workflow_factory: A factory function that creates a workflow instance for each request.
+            logger: The logger to use.
+            use_default_routers: Whether to use the default routers (chat, mount `data` and `output` directories).
+            env: The environment to run the server in.
+            include_ui: Whether to show an chat UI in the root path.
+            starter_questions: A list of starter questions to display in the chat UI.
+            server_url: The URL of the server.
+            api_prefix: The prefix for the API endpoints.
+            verbose: Whether to show verbose logs.
+        """
+        super().__init__(*args, **kwargs)
+
+        self.workflow_factory = workflow_factory
+        self.logger = logger or logging.getLogger("uvicorn")
+        self.verbose = verbose
+        self.include_ui = include_ui  # Store the explicitly passed value first
+        self.starter_questions = starter_questions
+        self.use_default_routers = use_default_routers or True
+
+        # Update the settings
+        if server_url:
+            server_settings.set_url(server_url)
+        if api_prefix:
+            server_settings.set_api_prefix(api_prefix)
+
+        if self.use_default_routers:
+            self.add_default_routers()
+
+        if str(env).lower() == "dev":
+            self.allow_cors("*")
+            if self.include_ui is None:
+                self.include_ui = True
+        if self.include_ui is None:
+            self.include_ui = False
+
+        if self.include_ui:
+            self.mount_ui()
+
+    @property
+    def _ui_config(self) -> dict:
+        config = {
+            "CHAT_API": f"{server_settings.api_url}/chat",
+            "STARTER_QUESTIONS": self.starter_questions,
+        }
+        is_llamacloud_configured = os.getenv("LLAMA_CLOUD_API_KEY") is not None
+        if is_llamacloud_configured:
+            config["LLAMA_CLOUD_API"] = (
+                f"{server_settings.api_url}/chat/config/llamacloud"
+            )
+        return config
+
+    # Default routers
+    def add_default_routers(self) -> None:
+        self.add_chat_router()
+        self.mount_data_dir()
+        self.mount_output_dir()
+
+    def add_chat_router(self) -> None:
+        """
+        Add the chat router.
+        """
+        self.include_router(
+            chat_router(
+                self.workflow_factory,
+                self.logger,
+            ),
+            prefix=server_settings.api_prefix,
+        )
+
+    def mount_ui(self) -> None:
+        """
+        Mount the UI.
+        """
+        # Check if the static folder exists
+        if self.include_ui:
+            if not os.path.exists(self.ui_path):
+                self.logger.warning(
+                    f"UI files not found, downloading UI to {self.ui_path}"
+                )
+                download_chat_ui(logger=self.logger, target_path=self.ui_path)
+            self._mount_static_files(directory=self.ui_path, path="/", html=True)
+            self._override_ui_config()
+
+    def _override_ui_config(self) -> None:
+        """
+        Override the UI config by writing a complete configuration file.
+        """
+        try:
+            config_path = os.path.join(self.ui_path, "config.js")
+            if not os.path.exists(config_path):
+                self.logger.error("Config file not found")
+                return
+            config_content = (
+                f"window.LLAMAINDEX = {json.dumps(self._ui_config, indent=2)};"
+            )
+            with open(config_path, "w") as f:
+                f.write(config_content)
+        except Exception as e:
+            self.logger.error(f"Error overriding UI config: {e}")
+
+    def mount_data_dir(self, data_dir: str = "data") -> None:
+        """
+        Mount the data directory.
+        """
+        self._mount_static_files(
+            directory=data_dir,
+            path=f"{server_settings.api_prefix}/files/data",
+            html=True,
+        )
+
+    def mount_output_dir(self, output_dir: str = "output") -> None:
+        """
+        Mount the output directory.
+        """
+        self._mount_static_files(
+            directory=output_dir,
+            path=f"{server_settings.api_prefix}/files/output",
+            html=True,
+        )
+
+    def _mount_static_files(
+        self, directory: str, path: str, html: bool = False
+    ) -> None:
+        """
+        Mount static files from a directory if it exists.
+        """
+        if os.path.exists(directory):
+            self.logger.info(f"Mounting static files '{directory}' at '{path}'")
+            self.mount(
+                path,
+                StaticFiles(directory=directory, check_dir=False, html=html),
+                name=f"{directory}-static",
+            )
+
+    def allow_cors(self, origin: str = "*") -> None:
+        """
+        Allow CORS for a specific origin.
+        """
+        self.add_middleware(
+            CORSMiddleware,
+            allow_origins=[origin],
+            allow_credentials=True,
+            allow_methods=["*"],
+            allow_headers=["*"],
+        )
@@ -149,8 +149,7 @@ def _create_index(
                    "type": "OPENAI_EMBEDDING",
                    "component": {
                        "api_key": os.getenv("OPENAI_API_KEY"),  # editable
-                        "model_name": Settings.embed_model.model_name
-                        or "text-embedding-3-small",
+                        "model_name": os.getenv("EMBEDDING_MODEL"),
                    },
                },
                "transform_config": {
@@ -11,10 +11,6 @@ class ServerSettings(BaseSettings):
        default="/api",
        description="The prefix for the API endpoints",
    )
-    workflow_factory_signature: str = Field(
-        default="",
-        description="The signature of the workflow factory function",
-    )

    @property
    def file_server_url_prefix(self) -> str:
@@ -44,9 +40,6 @@ class ServerSettings(BaseSettings):
        self.api_prefix = v
        self.validate_api_prefix(v)  # type: ignore

-    def set_workflow_factory(self, v: str) -> None:
-        self.workflow_factory_signature = v
-
    class Config:
        env_file_encoding = "utf-8"

@@ -0,0 +1,64 @@
+[build-system]
+build-backend = "poetry.core.masonry.api"
+requires = ["poetry-core"]
+
+[tool.codespell]
+check-filenames = true
+check-hidden = true
+# Feel free to un-skip examples, and experimental, you will just need to
+# work through many typos (--write-changes and --interactive will help)
+skip = "*.csv,*.html,*.json,*.jsonl,*.pdf,*.txt,*.ipynb"
+
+[tool.mypy]
+disallow_untyped_defs = true
+# Remove venv skip when integrated with pre-commit
+exclude = ["_static", "build", "examples", "notebooks", "venv"]
+ignore_missing_imports = true
+namespace_packages = true
+explicit_package_bases = true
+python_version = "3.10"
+
+[tool.poetry]
+authors = ["Your Name <you@example.com>"]
+description = "llama-index fastapi server"
+exclude = ["**/BUILD"]
+license = "MIT"
+name = "llama-index-server"
+packages = [{include = "llama_index/"}]
+readme = "README.md"
+version = "0.1.6"
+
+[tool.poetry.dependencies]
+python = ">=3.9,<4.0"
+fastapi = {extras = ["standard"], version = "^0.115.11"}
+cachetools = "^5.5.2"
+requests = "^2.32.3"
+pydantic-settings = "^2.8.1"
+llama-index-core = "0.12.25"
+llama-index-readers-file = "^0.4.6"
+llama-index-indices-managed-llama-cloud = "0.6.3"
+
+[tool.poetry.group.dev.dependencies]
+black = {extras = ["jupyter"], version = "<=23.9.1,>=23.7.0"}
+codespell = {extras = ["toml"], version = ">=v2.2.6"}
+e2b-code-interpreter = "^1.1.1"
+ipython = "8.10.0"
+jupyter = "^1.0.0"
+markdown = "^3.7"
+mypy = "1.15.0"
+pre-commit = "3.2.0"
+pylint = "2.15.10"
+pytest = "^8.3.5"
+pytest-asyncio = "^0.25.3"
+pytest-mock = "3.11.1"
+ruff = "0.0.292"
+tree-sitter-languages = "^1.8.0"
+types-Deprecated = ">=0.1.0"
+types-PyYAML = "^6.0.12.12"
+types-protobuf = "^4.24.0.4"
+types-redis = "4.5.5.0"
+types-requests = "2.28.11.8"  # TODO: unpin when mypy>0.991
+types-setuptools = "67.1.0.0"
+xhtml2pdf = "^0.2.17"
+pytest-cov = "^6.0.0"
+llama-cloud = "^0.1.17"
@@ -1,12 +1,10 @@
 import asyncio
 import logging
-from typing import Any, AsyncGenerator
 from unittest.mock import AsyncMock, MagicMock

 import pytest

 from llama_index.core.agent.workflow.workflow_events import AgentStream
-from llama_index.core.types import MessageRole
 from llama_index.core.workflow import StopEvent
 from llama_index.core.workflow.handler import WorkflowHandler
 from llama_index.server.api.models import ChatAPIMessage, ChatRequest
@@ -15,19 +13,17 @@ from llama_index.server.api.utils.vercel_stream import VercelStreamResponse


@pytest.fixture()
-def logger() -> logging.Logger:
+def logger():
    return logging.getLogger("test")


@pytest.fixture()
-def chat_request() -> ChatRequest:
-    return ChatRequest(
-        messages=[ChatAPIMessage(role=MessageRole.USER, content="test message")]
-    )
+def chat_request():
+    return ChatRequest(messages=[ChatAPIMessage(role="user", content="test message")])


@pytest.fixture()
-def mock_workflow_handler() -> AsyncMock:
+def mock_workflow_handler():
    handler = AsyncMock(spec=WorkflowHandler)
    handler.accumulate_text = MagicMock()
    return handler
@@ -36,11 +32,8 @@ def mock_workflow_handler() -> AsyncMock:
 class TestEventStream:
    @pytest.mark.asyncio()
    async def test_stream_content_with_agent_stream(
-        self,
-        mock_workflow_handler: AsyncMock,
-        chat_request: ChatRequest,
-        logger: logging.Logger,
-    ) -> None:
+        self, mock_workflow_handler, chat_request, logger
+    ):
        # Setup
        mock_workflow_handler.stream_events.return_value = (
            self._mock_agent_stream_events()
@@ -55,17 +48,15 @@ class TestEventStream:
        ]

        # Assert
-        assert len(result) == 2
-        assert result[0] == VercelStreamResponse.convert_text("Hello")
-        assert result[1] == VercelStreamResponse.convert_text(" World")
+        assert len(result) == 3  # Empty start + 2 text chunks
+        assert result[0] == VercelStreamResponse.convert_text("")
+        assert result[1] == VercelStreamResponse.convert_text("Hello")
+        assert result[2] == VercelStreamResponse.convert_text(" World")

    @pytest.mark.asyncio()
    async def test_stream_content_with_stop_event_string(
-        self,
-        mock_workflow_handler: AsyncMock,
-        chat_request: ChatRequest,
-        logger: logging.Logger,
-    ) -> None:
+        self, mock_workflow_handler, chat_request, logger
+    ):
        # Setup
        mock_workflow_handler.stream_events.return_value = (
            self._mock_stop_event_string()
@@ -80,16 +71,14 @@ class TestEventStream:
        ]

        # Assert
-        assert len(result) == 1
-        assert result[0] == VercelStreamResponse.convert_text("Final answer")
+        assert len(result) == 2  # Empty start + result string
+        assert result[0] == VercelStreamResponse.convert_text("")
+        assert result[1] == VercelStreamResponse.convert_text("Final answer")

    @pytest.mark.asyncio()
    async def test_stream_content_with_stop_event_delta_objects(
-        self,
-        mock_workflow_handler: AsyncMock,
-        chat_request: ChatRequest,
-        logger: logging.Logger,
-    ) -> None:
+        self, mock_workflow_handler, chat_request, logger
+    ):
        # Setup
        mock_workflow_handler.stream_events.return_value = (
            self._mock_stop_event_delta_objects()
@@ -104,17 +93,15 @@ class TestEventStream:
        ]

        # Assert
-        assert len(result) == 2
-        assert result[0] == VercelStreamResponse.convert_text("Delta 1")
-        assert result[1] == VercelStreamResponse.convert_text("Delta 2")
+        assert len(result) == 3  # Empty start + 2 delta chunks
+        assert result[0] == VercelStreamResponse.convert_text("")
+        assert result[1] == VercelStreamResponse.convert_text("Delta 1")
+        assert result[2] == VercelStreamResponse.convert_text("Delta 2")

    @pytest.mark.asyncio()
    async def test_stream_content_with_event_with_to_response(
-        self,
-        mock_workflow_handler: AsyncMock,
-        chat_request: ChatRequest,
-        logger: logging.Logger,
-    ) -> None:
+        self, mock_workflow_handler, chat_request, logger
+    ):
        # Setup
        mock_workflow_handler.stream_events.return_value = (
            self._mock_event_with_to_response()
@@ -129,16 +116,14 @@ class TestEventStream:
        ]

        # Assert
-        assert len(result) == 1
-        assert result[0] == VercelStreamResponse.convert_data({"event_type": "test"})
+        assert len(result) == 2  # Empty start + event with to_response
+        assert result[0] == VercelStreamResponse.convert_text("")
+        assert result[1] == VercelStreamResponse.convert_data({"event_type": "test"})

    @pytest.mark.asyncio()
    async def test_stream_content_with_event_with_model_dump(
-        self,
-        mock_workflow_handler: AsyncMock,
-        chat_request: ChatRequest,
-        logger: logging.Logger,
-    ) -> None:
+        self, mock_workflow_handler, chat_request, logger
+    ):
        # Setup
        mock_workflow_handler.stream_events.return_value = (
            self._mock_event_with_model_dump()
@@ -153,19 +138,17 @@ class TestEventStream:
        ]

        # Assert
-        assert len(result) == 1
-        assert result[0] == VercelStreamResponse.convert_data(None)  # type: ignore
+        assert len(result) == 2  # Empty start + event with model_dump
+        assert result[0] == VercelStreamResponse.convert_text("")
+        assert result[1] == VercelStreamResponse.convert_data(None)

    @pytest.mark.asyncio()
    async def test_stream_content_with_cancelled_error(
-        self,
-        mock_workflow_handler: AsyncMock,
-        chat_request: ChatRequest,
-        logger: logging.Logger,
-    ) -> None:
+        self, mock_workflow_handler, chat_request, logger
+    ):
        # Setup
        mock_workflow_handler.stream_events.side_effect = asyncio.CancelledError()
-        logger.warning = MagicMock()  # type: ignore
+        logger.warning = MagicMock()

        # Execute
        result = [
@@ -182,15 +165,12 @@ class TestEventStream:

    @pytest.mark.asyncio()
    async def test_stream_content_with_exception(
-        self,
-        mock_workflow_handler: AsyncMock,
-        chat_request: ChatRequest,
-        logger: logging.Logger,
-    ) -> None:
+        self, mock_workflow_handler, chat_request, logger
+    ):
        # Setup
        error_message = "Test error"
        mock_workflow_handler.stream_events.side_effect = Exception(error_message)
-        logger.error = MagicMock()  # type: ignore
+        logger.error = MagicMock()

        # Execute
        result = [
@@ -206,7 +186,7 @@ class TestEventStream:
        mock_workflow_handler.cancel_run.assert_called_once()
        logger.error.assert_called_once()

-    async def _mock_agent_stream_events(self) -> AsyncGenerator[AgentStream, Any]:
+    async def _mock_agent_stream_events(self):
        yield AgentStream(
            delta="Hello", response="", current_agent_name="", tool_calls=[], raw=""
        )
@@ -214,9 +194,7 @@ class TestEventStream:
            delta=" World", response="", current_agent_name="", tool_calls=[], raw=""
        )

-    async def _mock_agent_stream_with_empty_deltas(
-        self,
-    ) -> AsyncGenerator[AgentStream, Any]:
+    async def _mock_agent_stream_with_empty_deltas(self):
        yield AgentStream(
            delta="   ",  # Empty delta with spaces - should be filtered
            response="",
@@ -239,14 +217,14 @@ class TestEventStream:
            raw="",
        )

-    async def _mock_stop_event_string(self) -> AsyncGenerator[StopEvent, Any]:
+    async def _mock_stop_event_string(self):
        yield StopEvent(result="Final answer")

-    async def _mock_stop_event_delta_objects(self) -> AsyncGenerator[StopEvent, Any]:
-        async def generator() -> AsyncGenerator[Any, Any]:
+    async def _mock_stop_event_delta_objects(self):
+        async def generator():
            # Create proper objects with delta attribute that can be serialized
            class ObjectWithDelta:
-                def __init__(self, delta_value: str) -> None:
+                def __init__(self, delta_value) -> None:
                    self.delta = delta_value

            yield ObjectWithDelta("Delta 1")
@@ -254,15 +232,15 @@ class TestEventStream:

        yield StopEvent(result=generator())

-    async def _mock_dict_event(self) -> AsyncGenerator[dict[Any, Any], Any]:
+    async def _mock_dict_event(self):
        yield {"key": "value"}

-    async def _mock_event_with_to_response(self) -> AsyncGenerator[Any, Any]:
+    async def _mock_event_with_to_response(self):
        event = MagicMock()
        event.to_response.return_value = {"event_type": "test"}
        yield event

-    async def _mock_event_with_model_dump(self) -> AsyncGenerator[Any, Any]:
+    async def _mock_event_with_model_dump(self):
        event = MagicMock()
        event.model_dump.return_value = {"name": "test_event"}
        # Override to_response to return None - this means convert_data(None) will be called
@@ -0,0 +1,106 @@
+import pytest
+from httpx import ASGITransport, AsyncClient
+
+from llama_index.core.agent.workflow import AgentWorkflow
+from llama_index.core.llms import MockLLM
+from llama_index.server import LlamaIndexServer
+
+
+def fetch_weather(city: str) -> str:
+    """Fetch the weather for a given city."""
+    return f"The weather in {city} is sunny."
+
+
+def _agent_workflow() -> AgentWorkflow:
+    # Use MockLLM instead of default OpenAI
+    mock_llm = MockLLM()
+    return AgentWorkflow.from_tools_or_functions(
+        tools_or_functions=[fetch_weather],
+        verbose=True,
+        llm=mock_llm,
+    )
+
+
+@pytest.fixture()
+def server() -> LlamaIndexServer:
+    """Fixture to create a LlamaIndexServer instance."""
+    return LlamaIndexServer(
+        workflow_factory=_agent_workflow,
+        verbose=True,
+        use_default_routers=True,
+        mount_ui=False,
+        env="dev",
+    )
+
+
+@pytest.mark.asyncio()
+async def test_server_has_chat_route(server: LlamaIndexServer) -> None:
+    """Test that the server has the chat API route."""
+    chat_route_exists = any(route.path == "/api/chat" for route in server.routes)
+    assert chat_route_exists, "Chat API route not found in server routes"
+
+
+@pytest.mark.asyncio()
+async def test_server_swagger_docs(server: LlamaIndexServer) -> None:
+    """Test that the server serves Swagger UI docs."""
+    async with AsyncClient(
+        transport=ASGITransport(app=server), base_url="http://test"
+    ) as ac:
+        response = await ac.get("/docs")
+        assert response.status_code == 200
+        assert "text/html" in response.headers["content-type"]
+        assert "Swagger UI" in response.text
+
+
+@pytest.mark.asyncio()
+async def test_ui_is_downloaded(server: LlamaIndexServer) -> None:
+    """
+    Test if the UI is downloaded and mounted correctly.
+    """
+    import os
+    import shutil
+
+    # Clean up any existing static directory first
+    if os.path.exists(".ui"):
+        shutil.rmtree(".ui")
+
+    # Create a new server with UI enabled
+    ui_server = LlamaIndexServer(
+        workflow_factory=_agent_workflow,
+        verbose=True,
+        use_default_routers=True,
+        env="dev",
+        include_ui=True,
+    )
+
+    # Verify that static directory was created with index.html
+    assert os.path.exists("./.ui"), "Static directory was not created"
+    assert os.path.isdir("./.ui"), "Static path is not a directory"
+    assert os.path.exists("./.ui/index.html"), "index.html was not downloaded"
+
+    # Check if the UI is mounted and accessible
+    async with AsyncClient(
+        transport=ASGITransport(app=ui_server), base_url="http://test"
+    ) as ac:
+        response = await ac.get("/")
+        assert response.status_code == 200
+        assert "text/html" in response.headers["content-type"]
+
+    # Clean up after test
+    shutil.rmtree("./.ui")
+
+
+@pytest.mark.asyncio()
+async def test_ui_is_accessible(server: LlamaIndexServer) -> None:
+    """
+    Test if the UI is accessible.
+    """
+    # Manually trigger UI mounting
+    server.mount_ui()
+
+    async with AsyncClient(
+        transport=ASGITransport(app=server), base_url="http://test"
+    ) as ac:
+        response = await ac.get("/")
+        assert response.status_code == 200
+        assert "text/html" in response.headers["content-type"]
@@ -1,52 +1,83 @@
 {
-  "name": "create-llama-monorepo",
-  "version": "1.0.0",
-  "private": true,
-  "description": "Monorepo for create-llama",
+  "name": "create-llama",
+  "version": "0.4.0",
+  "description": "Create LlamaIndex-powered apps with one command",
  "keywords": [
    "rag",
-    "llamaindex"
+    "llamaindex",
+    "next.js"
  ],
  "repository": {
    "type": "git",
-    "url": "https://github.com/run-llama/create-llama"
+    "url": "https://github.com/run-llama/create-llama",
+    "directory": "packages/create-llama"
  },
  "license": "MIT",
-  "workspaces": [
-    "packages/*"
+  "bin": {
+    "create-llama": "./dist/index.js"
+  },
+  "files": [
+    "dist"
  ],
  "scripts": {
-    "dev": "pnpm -r dev",
-    "build": "pnpm -r build",
-    "e2e": "pnpm -r e2e",
-    "lint": "eslint .",
+    "build": "bash ./scripts/build.sh",
+    "build:ncc": "pnpm run clean && ncc build ./index.ts -o ./dist/ --minify --no-cache --no-source-map-register",
+    "clean": "rimraf --glob ./dist ./templates/**/__pycache__ ./templates/**/node_modules ./templates/**/poetry.lock",
+    "dev": "ncc build ./index.ts -w -o dist/",
+    "e2e": "playwright test",
+    "e2e:python": "playwright test e2e/shared e2e/python",
+    "e2e:typescript": "playwright test e2e/shared e2e/typescript",
    "format": "prettier --ignore-unknown --cache --check .",
    "format:write": "prettier --ignore-unknown --write .",
+    "lint": "eslint . --ignore-pattern dist --ignore-pattern e2e/cache",
+    "new-snapshot": "pnpm run build && changeset version --snapshot",
+    "new-version": "pnpm run build && changeset version",
+    "pack-install": "bash ./scripts/pack.sh",
    "prepare": "husky",
-    "new-snapshot": "pnpm -r build && changeset version --snapshot",
-    "new-version": "pnpm -r build && changeset version",
-    "release": "pnpm -r build && changeset publish",
-    "release-snapshot": "pnpm -r build && changeset publish --tag snapshot"
+    "release": "pnpm run build && changeset publish",
+    "release-snapshot": "pnpm run build && changeset publish --tag snapshot"
+  },
+  "dependencies": {
+    "@types/async-retry": "1.4.2",
+    "@types/ci-info": "2.0.0",
+    "@types/cross-spawn": "6.0.0",
+    "@types/fs-extra": "11.0.4",
+    "@types/node": "^20.11.7",
+    "@types/prompts": "2.4.2",
+    "@types/tar": "6.1.5",
+    "@types/validate-npm-package-name": "3.0.0",
+    "async-retry": "1.3.1",
+    "async-sema": "3.0.1",
+    "ci-info": "github:watson/ci-info#f43f6a1cefff47fb361c88cf4b943fdbcaafe540",
+    "commander": "12.1.0",
+    "cross-spawn": "7.0.3",
+    "fast-glob": "3.3.1",
+    "fs-extra": "11.2.0",
+    "global-agent": "^3.0.0",
+    "got": "10.7.0",
+    "ollama": "^0.5.0",
+    "ora": "^8.0.1",
+    "picocolors": "1.0.0",
+    "prompts": "2.4.2",
+    "smol-toml": "^1.1.4",
+    "tar": "6.1.15",
+    "terminal-link": "^3.0.0",
+    "update-check": "1.5.4",
+    "validate-npm-package-name": "3.0.0",
+    "yaml": "2.4.1"
  },
  "devDependencies": {
    "@changesets/cli": "^2.27.1",
-    "bunchee": "6.4.0",
+    "@playwright/test": "^1.41.1",
+    "@vercel/ncc": "0.38.1",
+    "eslint": "^8.56.0",
+    "eslint-config-prettier": "^8.10.0",
    "husky": "^9.0.10",
-    "lint-staged": "^15.2.11",
-    "typescript-eslint": "^8.18.0",
-    "globals": "^15.12.0",
-    "eslint": "9.22.0",
-    "@eslint/js": "^9.25.0",
-    "eslint-config-next": "^15.1.0",
-    "eslint-config-prettier": "^9.1.0",
-    "eslint-plugin-react": "7.37.2",
-    "prettier": "^3.4.2",
-    "prettier-plugin-organize-imports": "^4.1.0",
-    "prettier-plugin-tailwindcss": "^0.6.11",
-    "typescript": "^5.7.3",
-    "@types/node": "^22.9.0",
-    "@types/react": "^19",
-    "@types/react-dom": "^19"
+    "prettier": "^3.2.5",
+    "prettier-plugin-organize-imports": "^3.2.4",
+    "rimraf": "^5.0.5",
+    "typescript": "^5.3.3",
+    "wait-port": "^1.1.0"
  },
  "packageManager": "pnpm@9.0.5",
  "engines": {
@@ -1,65 +0,0 @@
-# See https://help.github.com/articles/ignoring-files/ for more about ignoring files.
-
-# dependencies
-node_modules
-.pnp
-.pnpm-store
-.pnp.js
-
-# testing
-coverage
-.coverage
-
-# next.js
-.next/
-out/
-build
-
-# misc
-.DS_Store
-*.pem
-
-# debug
-npm-debug.log*
-yarn-debug.log*
-yarn-error.log*
-
-# local env files
-.env
-.env.local
-.env.development.local
-.env.test.local
-.env.production.local
-
-# build
-dist/
-lib/
-
-# e2e
-.cache
-test-results/
-playwright-report/
-blob-report/
-playwright/.cache/
-.tsbuildinfo
-e2e/cache
-
-# intellij
-**/.idea
-
-# Python
-.mypy_cache/
-venv/
-.venv/
-dist/
-.__pycache__
-__pycache__
-.python-version
-.ui
-
-# build artifacts
-create-llama-*.tgz
-
-# copied from root
-README.md
-LICENSE.md
@@ -1,285 +0,0 @@
-import { expect, test } from "@playwright/test";
-import { exec } from "child_process";
-import fs from "fs";
-import path from "path";
-import util from "util";
-import { TemplateFramework, TemplateType, TemplateUseCase, TemplateVectorDB } from "../../helpers/types";
-import { RunCreateLlamaOptions, createTestDir, runCreateLlama } from "../utils";
-
-const execAsync = util.promisify(exec);
-
-const templateFramework: TemplateFramework = process.env.FRAMEWORK
-  ? (process.env.FRAMEWORK as TemplateFramework)
-  : "fastapi";
-const templateType: TemplateType = process.env.TEMPLATE_TYPE
-  ? (process.env.TEMPLATE_TYPE as TemplateType)
-  : "llamaindexserver";
-const useCases: TemplateUseCase[] = [
-  "agentic_rag",
-  "deep_research",
-  "financial_report",
-  "code_generator",
-  "document_generator",
-];
-const dataSource: string = process.env.DATASOURCE
-  ? process.env.DATASOURCE
-  : "--example-file";
-
-test.describe("Mypy check", () => {
-  test.describe.configure({ retries: 0 });
-
-    // Test for streaming template
-    test.describe("StreamingTemplate", () => {
-      test.skip(templateType !== "streaming", `skipping streaming test for ${templateType}`);
-      if (
-        dataSource === "--example-file" // XXX: this test provides its own data source - only trigger it on one data source (usually the CI matrix will trigger multiple data sources)
-      ) {
-        // vectorDBs, tools, and data source combinations to test
-        const vectorDbs: TemplateVectorDB[] = [
-          "mongo",
-          "pg",
-          "pinecone",
-          "milvus",
-          "astra",
-          "qdrant",
-          "chroma",
-          "weaviate",
-        ];
-        const toolOptions = [
-          "wikipedia.WikipediaToolSpec",
-          "google.GoogleSearchToolSpec",
-          "document_generator",
-          "artifact",
-        ];
-
-        const dataSources = [
-          "--example-file",
-          "--web-source https://www.example.com",
-          "--db-source mysql+pymysql://user:pass@localhost:3306/mydb",
-        ];
-
-        const observabilityOptions = ["llamatrace", "traceloop"];
-
-        // Test vector databases
-        for (const vectorDb of vectorDbs) {
-          test(`vectorDB: ${vectorDb} ${templateType}`, async () => {
-            const cwd = await createTestDir();
-            const { pyprojectPath } = await createAndCheckLlamaProject({
-              options: {
-                cwd,
-                templateType: "streaming",
-                templateFramework,
-                dataSource: "--example-file",
-                vectorDb,
-                tools: "none",
-                port: 3000,
-                postInstallAction: "none",
-                templateUI: undefined,
-                appType: "--no-frontend",
-                llamaCloudProjectName: undefined,
-                llamaCloudIndexName: undefined,
-                observability: undefined,
-              },
-            });
-
-            const pyprojectContent = fs.readFileSync(pyprojectPath, "utf-8");
-            if (vectorDb !== "none") {
-              if (vectorDb === "pg") {
-                expect(pyprojectContent).toContain(
-                  "llama-index-vector-stores-postgres",
-                );
-              } else {
-                expect(pyprojectContent).toContain(
-                  `llama-index-vector-stores-${vectorDb}`,
-                );
-              }
-            }
-          });
-        }
-
-        // // Test tools
-        for (const tool of toolOptions) {
-          test(`tool: ${tool} ${templateType}`, async () => {
-            const cwd = await createTestDir();
-            const { pyprojectPath } = await createAndCheckLlamaProject({
-              options: {
-                cwd,
-                templateType: "streaming",
-                templateFramework,
-                dataSource: "--example-file",
-                vectorDb: "none",
-                tools: tool,
-                port: 3000,
-                postInstallAction: "none",
-                templateUI: undefined,
-                appType: "--no-frontend",
-                llamaCloudProjectName: undefined,
-                llamaCloudIndexName: undefined,
-                observability: undefined,
-              },
-            });
-
-            const pyprojectContent = fs.readFileSync(pyprojectPath, "utf-8");
-            if (tool === "wikipedia.WikipediaToolSpec") {
-              expect(pyprojectContent).toContain("wikipedia");
-            }
-            if (tool === "google.GoogleSearchToolSpec") {
-              expect(pyprojectContent).toContain("google");
-            }
-          });
-        }
-
-        // // Test data sources
-        for (const dataSource of dataSources) {
-          test(`data source: ${dataSource} ${templateType}`, async () => {
-          const dataSourceType = dataSource.split(" ")[0];
-            const cwd = await createTestDir();
-            const { pyprojectPath } = await createAndCheckLlamaProject({
-              options: {
-                cwd,
-                templateType: "streaming",
-                templateFramework,
-                dataSource,
-                vectorDb: "none",
-                tools: "none",
-                port: 3000,
-                postInstallAction: "none",
-                templateUI: undefined,
-                appType: "--no-frontend",
-                llamaCloudProjectName: undefined,
-                llamaCloudIndexName: undefined,
-                observability: undefined,
-              },
-            });
-
-            const pyprojectContent = fs.readFileSync(pyprojectPath, "utf-8");
-            if (dataSource.includes("--web-source")) {
-              expect(pyprojectContent).toContain("llama-index-readers-web");
-            }
-            if (dataSource.includes("--db-source")) {
-              expect(pyprojectContent).toContain("llama-index-readers-database");
-            }
-          });
-        }
-
-        // Test observability options
-        for (const observability of observabilityOptions) {
-          test.describe(`observability: ${observability} ${templateType}`, async () => {
-            const cwd = await createTestDir();
-
-            const { pyprojectPath } = await createAndCheckLlamaProject({
-              options: {
-                cwd,
-                templateType: "streaming",
-                templateFramework,
-                dataSource: "--example-file",
-                vectorDb: "none",
-                tools: "none",
-                port: 3000,
-                postInstallAction: "none",
-                templateUI: undefined,
-                appType: "--no-frontend",
-                llamaCloudProjectName: undefined,
-                llamaCloudIndexName: undefined,
-                observability,
-              },
-            });
-          });
-        }
-      }
-    });
-
-    test.describe("LlamaIndexServer", async () => {
-      test.skip(templateType !== "llamaindexserver", `skipping llamaindexserver test for ${templateType}`);
-      test.skip(dataSource !== "--example-file", `skipping llamaindexserver test for ${dataSource}`);
-      for (const useCase of useCases) {
-        const cwd = await createTestDir();
-        await createAndCheckLlamaProject({
-          options: {
-            cwd,
-            templateType: "llamaindexserver",
-            templateFramework,
-            dataSource,
-            vectorDb: "none",
-            tools: "none",
-            port: 3000,
-            postInstallAction: "none",
-            templateUI: undefined,
-            appType: "--no-frontend",
-            llamaCloudProjectName: undefined,
-            llamaCloudIndexName: undefined,
-            observability: undefined,
-            useCase,
-          },
-        });
-      }
-    });
-
-    async function createAndCheckLlamaProject({
-      options,
-    }: {
-      options: RunCreateLlamaOptions;
-    }): Promise<{ pyprojectPath: string; projectPath: string }> {
-      const result = await runCreateLlama(options);
-      const name = result.projectName;
-      const projectPath = path.join(options.cwd, name);
-
-      // Check if the app folder exists
-      expect(fs.existsSync(projectPath)).toBeTruthy();
-
-      // Check if pyproject.toml exists
-      const pyprojectPath = path.join(projectPath, "pyproject.toml");
-      expect(fs.existsSync(pyprojectPath)).toBeTruthy();
-
-      // Modify environment for the command
-      const commandEnv = {
-        ...process.env,
-      };
-
-      console.log("Running uv venv...");
-      try {
-        const { stdout: venvStdout, stderr: venvStderr } = await execAsync(
-          "uv venv",
-          { cwd: projectPath, env: commandEnv },
-        );
-        console.log("uv venv stdout:", venvStdout);
-        console.error("uv venv stderr:", venvStderr);
-      } catch (error) {
-        console.error("Error running uv venv:", error);
-        throw error; // Re-throw error to fail the test
-      }
-
-      console.log("Running uv sync...");
-      try {
-        const { stdout: syncStdout, stderr: syncStderr } = await execAsync(
-          "uv sync --all-extras",
-          { cwd: projectPath, env: commandEnv },
-        );
-        console.log("uv sync stdout:", syncStdout);
-        console.error("uv sync stderr:", syncStderr);
-      } catch (error) {
-        console.error("Error running uv sync:", error);
-        throw error; // Re-throw error to fail the test
-      }
-
-      console.log("Running uv run mypy ....");
-      try {
-        const { stdout: mypyStdout, stderr: mypyStderr } = await execAsync(
-          "uv run mypy .",
-          { cwd: projectPath, env: commandEnv },
-        );
-        console.log("uv run mypy stdout:", mypyStdout);
-        console.error("uv run mypy stderr:", mypyStderr);
-        // Assuming mypy success means no output or specific success message
-        // Adjust checks based on actual expected mypy output
-      } catch (error) {
-        console.error("Error running mypy:", error);
-        throw error;
-      }
-
-      // If we reach this point without throwing an error, the test passes
-      expect(true).toBeTruthy();
-
-    return { pyprojectPath, projectPath };
-  }
-});
--- a/Show More
+++ b/Show More
Author	SHA1	Message	Date
leehuwuj	670e2631f3	bump @llamaindex/server	2025-04-02 17:20:32 +07:00
leehuwuj	afb519d231	fix llamacloud api and markdown issue	2025-04-02 16:51:21 +07:00