mirror of
https://github.com/Significant-Gravitas/AutoGPT.git
synced 2026-11-09 17:53:56 -10:00
Nicholas Tindle
fa70e6436c
### Why / What / How
Several manually authored guides had broken relative links after
documentation assets moved, stale model lists after catalog changes,
outdated ChatGPT connection instructions, and an incomplete External API
scope table. This PR updates those guides against the current `dev`
source, catalog, routes, and UI tests without touching generator-owned
integration output.
### Changes 🏗️
- Repair 19 AI/ML API and Ollama image references plus the legacy docs
landing-page link.
- Point the External API guide at the existing OAuth guide and document
`WRITE_GRAPH` and `WRITE_LIBRARY`.
- Align AI/ML API, Gemini, and Ollama model lists with the current LLM
catalog.
- Align the ChatGPT/Codex preview steps with the current OpenAI
integration labels, visible model tiers, and server-selected execution
path.
### Agents and large language models used
- Codex with GPT-5
### Checklist 📋
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
- [x] `poetry run python scripts/generate_block_docs.py --check`
- [x] Full local Markdown target, exact-case, anchor, fence, and
GitBook-directive scan
- [x] `bash autogpt_platform/installer/test-single-container-handoff.sh`
- [x] `backend/copilot/codex_model_router_test.py` (8 passed)
- [x] `git diff --check origin/dev...HEAD`
- [x] Repository pre-commit hooks for `origin/dev...HEAD`
- [ ] `backend/copilot/provider_tiers_test.py` (the local
`SpinTestServer` fixture stalled and was stopped after three minutes; no
assertion failure was emitted)
<details>
<summary>Example test plan</summary>
- [ ] Create from scratch and execute an agent with at least 3 blocks
- [ ] Import an agent from file upload, and confirm it executes
correctly
- [ ] Upload agent to marketplace
- [ ] Import an agent from marketplace and confirm it executes correctly
- [ ] Edit an agent from monitor, and confirm it executes correctly
</details>
#### For configuration changes:
- [ ] `.env.default` is updated or already compatible with my changes
- [ ] `docker-compose.yml` is updated or already compatible with my
changes
- [ ] I have included a list of my configuration changes in the PR
description (under **Changes**)
<details>
<summary>Examples of configuration changes</summary>
- Changing ports
- Adding new services that need to communicate with each other
- Secrets or environment variable changes
- New or infrastructure changes such as databases
</details>
<!-- CURSOR_SUMMARY -->
---
> [!NOTE]
> **Low Risk**
> Markdown and link fixes only; no runtime, auth, or API behavior
changes.
>
> **Overview**
> **Documentation-only refresh** for platform guides that had broken
assets, stale catalog text, and outdated integration UI wording.
>
> Relative links and screenshots now point at `../content/imgs/...`
(AI/ML API, Ollama) and the docs landing page uses
`../platform/new_blocks.md`. The External API guide links to
`oauth-guide.md` instead of the removed SSO guide, and documents
**`WRITE_GRAPH`** and **`WRITE_LIBRARY`** scopes.
>
> **Model tables** in the AI/ML API, Gemini, and Ollama guides match the
current LLM catalog (including a slimmer AI/ML API list and updated
Gemini IDs); Ollama adds **`llama3.3`** to the default compatible list.
>
> The **ChatGPT/Codex subscription preview** doc matches current UI:
connect via **Settings → Integrations → OpenAI → ChatGPT → Sign in with
ChatGPT**, AutoPilot tier controls are **Balanced/Advanced**, and Codex
routing is explained with an execution-path × tier mapping table instead
of the older Fast/Thinking prose.
>
> <sup>Reviewed by [Cursor Bugbot](https://cursor.com/bugbot) for commit
0d678fe0e1. Bugbot is set up for automated
code reviews on this repo. Configure
[here](https://www.cursor.com/dashboard/bugbot).</sup>
<!-- /CURSOR_SUMMARY -->
3.8 KiB
3.8 KiB
🧠 Running AI/ML API with AutoGPT
Follow these steps to connect AI/ML API with the AutoGPT platform for high-performance AI text generation.
✅ Prerequisites
- Make sure you have gone through and completed the AutoGPT Setup Guide, and AutoGPT is running locally at
http://localhost:3000. - You have an API key from AI/ML API.
⚙️ Setup Steps
1. Start AutoGPT Locally
Follow the official guide: 📖 AutoGPT Getting Started Guide
Make sure AutoGPT is running and accessible at: http://localhost:3000
💡 Keep AutoGPT running in a terminal or Docker throughout the session.
2. Open the Visual Builder
Open your browser and go to: http://localhost:3000/build
Or click “Build” in the navigation bar.
3. Add an AI Text Generator Block
- Click the "Blocks" button on the left sidebar.
- In the search bar, type
AI Text Generator. - Drag the block into the canvas.
4. Select an AI/ML API Model
Click the AI Text Generator block to configure it.
In the LLM Model dropdown, select the model currently available from AI/ML API:
| Model | Model ID |
|---|---|
| Llama 3.3 70B Instruct Turbo | meta-llama/Llama-3.3-70B-Instruct-Turbo |
✅ This model is available via the OpenAI-compatible API from AI/ML API
5. Configure the Prompt and API Key
Inside the AI Text Generator block:
- Enter your prompt text in the Prompt field.
- Enter your AI/ML API Key in the designated field.
🔐 You can get your key from: https://aimlapi.com/app/keys/
6. Save Your Agent
Click the “Save” button at the top-right of the builder interface:
- Give your agent a name (e.g.,
aimlapi_test_agent). - Click “Save Agent” to confirm.
💡 Saving allows reuse, scheduling, and chaining in larger workflows.
7. Run Your Agent
From the workspace:
- Press “Run” next to your saved agent.
- The request will be sent to the selected AI/ML API model.
8. View the Output
- Scroll to the AI Text Generator block.
- Check the Output panel below it.
- You can copy, export, or pass the result to further blocks.
🔄 Expand Your Agent
Now that AI/ML API is connected, expand your workflow by chaining additional blocks:
- 🔧 Tools – fetch URLs, call APIs, scrape data
- 🧠 Memory – retain context across interactions
- ⚙️ Actions / Chains – create full pipelines
🎉 You’re now generating AI responses using enterprise-grade models from AI/ML API in AutoGPT!