fix: version

# Conflicts:
#	packages/core/package.json
#	pnpm-lock.yaml

.github/workflows/test.yml

@@ -85,6 +85,7 @@ jobs:
           if-no-files-found: error
   e2e-core-examples:
     strategy:
+      fail-fast: false
       matrix:
         packages:
           - cloudflare-worker-agent

examples/package.json

@@ -4,22 +4,22 @@
   "version": "0.0.4",
   "dependencies": {
     "@aws-crypto/sha256-js": "^5.2.0",
-    "@datastax/astra-db-ts": "^1.1.0",
+    "@datastax/astra-db-ts": "^1.2.1",
     "@notionhq/client": "^2.2.15",
-    "@pinecone-database/pinecone": "^2.2.0",
+    "@pinecone-database/pinecone": "^2.2.2",
     "@zilliz/milvus2-sdk-node": "^2.4.2",
-    "chromadb": "^1.7.3",
-    "commander": "^12.0.0",
+    "chromadb": "^1.8.1",
+    "commander": "^12.1.0",
     "dotenv": "^16.4.5",
-    "js-tiktoken": "^1.0.11",
-    "llamaindex": "*",
-    "mongodb": "^6.6.1",
+    "js-tiktoken": "^1.0.12",
+    "llamaindex": "^0.3.14",
+    "mongodb": "^6.7.0",
     "pathe": "^1.1.2"
   },
   "devDependencies": {
-    "@types/node": "^20.12.11",
+    "@types/node": "^20.14.1",
     "ts-node": "^10.9.2",
-    "tsx": "^4.9.3",
+    "tsx": "^4.11.2",
     "typescript": "^5.4.5"
   },
   "scripts": {

package.json

@@ -19,15 +19,15 @@
     "new-snapshot": "pnpm run build:release && changeset version --snapshot"
   },
   "devDependencies": {
-    "@changesets/cli": "^2.27.1",
-    "@typescript-eslint/eslint-plugin": "^7.8.0",
+    "@changesets/cli": "^2.27.5",
+    "@typescript-eslint/eslint-plugin": "^7.11.0",
     "eslint": "^8.57.0",
     "eslint-config-next": "^14.2.3",
     "eslint-config-prettier": "^9.1.0",
     "eslint-config-turbo": "^1.13.3",
     "eslint-plugin-react": "7.34.1",
     "husky": "^9.0.11",
-    "lint-staged": "^15.2.2",
+    "lint-staged": "^15.2.5",
     "madge": "^7.0.0",
     "prettier": "^3.2.5",
     "prettier-plugin-organize-imports": "^3.2.4",

packages/core/package.json

@@ -21,45 +21,45 @@
     "llamaindex"
   ],
   "dependencies": {
-    "@anthropic-ai/sdk": "^0.20.9",
+    "@anthropic-ai/sdk": "^0.21.1",
     "@aws-crypto/sha256-js": "^5.2.0",
-    "@datastax/astra-db-ts": "^1.1.0",
-    "@google-cloud/vertexai": "^1.1.0",
-    "@google/generative-ai": "^0.11.0",
-    "@grpc/grpc-js": "^1.10.7",
-    "@huggingface/inference": "^2.6.7",
+    "@datastax/astra-db-ts": "^1.2.1",
+    "@google-cloud/vertexai": "^1.2.0",
+    "@google/generative-ai": "^0.12.0",
+    "@grpc/grpc-js": "^1.10.8",
+    "@huggingface/inference": "^2.7.0",
     "@llamaindex/cloud": "0.0.5",
     "@llamaindex/env": "workspace:*",
-    "@mistralai/mistralai": "^0.2.0",
-    "@pinecone-database/pinecone": "^2.2.0",
+    "@mistralai/mistralai": "^0.4.0",
+    "@pinecone-database/pinecone": "^2.2.2",
     "@qdrant/js-client-rest": "^1.9.0",
-    "@types/lodash": "^4.17.1",
+    "@types/lodash": "^4.17.4",
     "@types/papaparse": "^5.3.14",
     "@types/pg": "^8.11.6",
-    "@xenova/transformers": "^2.17.1",
+    "@xenova/transformers": "^2.17.2",
     "@zilliz/milvus2-sdk-node": "^2.4.2",
-    "ajv": "^8.13.0",
-    "assemblyai": "^4.4.2",
-    "chromadb": "~1.7.3",
-    "cohere-ai": "^7.9.5",
-    "js-tiktoken": "^1.0.11",
+    "ajv": "^8.16.0",
+    "assemblyai": "^4.4.5",
+    "chromadb": "1.8.1",
+    "cohere-ai": "7.9.5",
+    "js-tiktoken": "^1.0.12",
     "lodash": "^4.17.21",
     "magic-bytes.js": "^1.10.0",
     "mammoth": "^1.7.2",
     "md-utils-ts": "^2.0.0",
-    "mongodb": "^6.6.1",
+    "mongodb": "^6.7.0",
     "notion-md-crawler": "^1.0.0",
-    "openai": "^4.46.0",
+    "openai": "^4.48.1",
     "papaparse": "^5.4.1",
     "pathe": "^1.1.2",
-    "pdf2json": "3.0.5",
-    "pg": "^8.11.5",
+    "pdf2json": "3.1.3",
+    "pg": "^8.12.0",
     "pgvector": "^0.1.8",
     "portkey-ai": "^0.1.16",
     "rake-modified": "^1.0.8",
     "string-strip-html": "^13.4.8",
     "wikipedia": "^2.1.2",
-    "wink-nlp": "^2.2.2"
+    "wink-nlp": "^2.3.0"
   },
   "peerDependencies": {
     "@notionhq/client": "^2.2.15"
@@ -67,9 +67,9 @@
   "devDependencies": {
     "@notionhq/client": "^2.2.15",
     "@swc/cli": "^0.3.12",
-    "@swc/core": "^1.5.5",
+    "@swc/core": "^1.5.24",
     "concurrently": "^8.2.2",
-    "glob": "^10.3.12",
+    "glob": "^10.4.1",
     "typescript": "^5.4.5"
   },
   "engines": {

packages/core/src/llm/huggingface.ts

@@ -1,7 +1,4 @@
-import {
-  HfInference,
-  type Options as HfInferenceOptions,
-} from "@huggingface/inference";
+import { HfInference } from "@huggingface/inference";
 import type {
   PreTrainedModel,
   PreTrainedTokenizer,
@@ -20,12 +17,49 @@ import type {
 } from "./types.js";
 import { streamConverter, wrapLLMEvent } from "./utils.js";
 
+// TODO workaround issue with @huggingface/inference@2.7.0
+interface HfInferenceOptions {
+  /**
+   * (Default: true) Boolean. If a request 503s and wait_for_model is set to false, the request will be retried with the same parameters but with wait_for_model set to true.
+   */
+  retry_on_error?: boolean;
+  /**
+   * (Default: true). Boolean. There is a cache layer on Inference API (serverless) to speedup requests we have already seen. Most models can use those results as is as models are deterministic (meaning the results will be the same anyway). However if you use a non deterministic model, you can set this parameter to prevent the caching mechanism from being used resulting in a real new query.
+   */
+  use_cache?: boolean;
+  /**
+   * (Default: false). Boolean. Do not load the model if it's not already available.
+   */
+  dont_load_model?: boolean;
+  /**
+   * (Default: false). Boolean to use GPU instead of CPU for inference (requires Startup plan at least).
+   */
+  use_gpu?: boolean;
+  /**
+   * (Default: false) Boolean. If the model is not ready, wait for it instead of receiving 503. It limits the number of requests required to get your inference done. It is advised to only set this flag to true after receiving a 503 error as it will limit hanging in your application to known places.
+   */
+  wait_for_model?: boolean;
+  /**
+   * Custom fetch function to use instead of the default one, for example to use a proxy or edit headers.
+   */
+  fetch?: typeof fetch;
+  /**
+   * Abort Controller signal to use for request interruption.
+   */
+  signal?: AbortSignal;
+  /**
+   * (Default: "same-origin"). String | Boolean. Credentials to use for the request. If this is a string, it will be passed straight on. If it's a boolean, true will be "include" and false will not send credentials at all.
+   */
+  includeCredentials?: string | boolean;
+}
+
 const DEFAULT_PARAMS = {
   temperature: 0.1,
   topP: 1,
   maxTokens: undefined,
   contextWindow: 3900,
 };
+
 export type HFConfig = Partial<typeof DEFAULT_PARAMS> &
   HfInferenceOptions & {
     model: string;
@@ -139,7 +173,7 @@ export class HuggingFaceInferenceAPI extends BaseLLM {
       inputs: this.messagesToPrompt(params.messages),
       parameters: this.metadata,
     });
-    yield* streamConverter(stream, (chunk) => ({
+    yield* streamConverter(stream, (chunk: any) => ({
       delta: chunk.token.text,
       raw: chunk,
     }));

packages/core/src/readers/PDFReader.ts

@@ -21,7 +21,7 @@ export class PDFReader implements BaseReader {
 
 async function readPDF(data: Buffer): Promise<string[]> {
   const parser = await import("pdf2json").then(
-    ({ default: Pdfparser }) => new Pdfparser(null, 1),
+    ({ default: Pdfparser }) => new Pdfparser(null, true),
   );
   const text = await new Promise<string>((resolve, reject) => {
     parser.on("pdfParser_dataError", (error) => {