AutoGPT is the vision of accessible AI for everyone, to use and to build on. Our mission is to provide the tools, so that you can focus on what matters.
docs(platform): refresh manual guides and links (#14447)
Nicholas Tindlefa70e6436c08
diff --git a/docs/content/index.md b/docs/content/index.mdindex fae4f2e1e7..23766a09c1 100644--- a/docs/content/index.md+++ b/docs/content/index.md@@ -58,7 +58,7 @@ Blocks represent actions and are the building blocks of your workflows, includin- Custom scripts or functions- Conditional logic and decision-making components-You can learn more under: [Build your own Blocks](platform/new_blocks.md)+You can learn more under: [Build your own Blocks](../platform/new_blocks.md)## Available Language Models@@ -84,4 +84,4 @@ This strategy allows us to share previously closed-source components, fostering- Read the [Getting Started docs](https://agpt.co/docs/platform/getting-started) to self-host- [Join the waitlist](https://agpt.co/waitlist) for the cloud-hosted beta-- [Contribute](contribute/index.md)\ No newline at end of file+- [Contribute](contribute/index.md)diff --git a/docs/platform/aimlapi.md b/docs/platform/aimlapi.mdindex 2ce2c26c2c..960bf2b01e 100644--- a/docs/platform/aimlapi.md+++ b/docs/platform/aimlapi.md@@ -24,7 +24,7 @@ Make sure AutoGPT is running and accessible at:> 💡 Keep AutoGPT running in a terminal or Docker throughout the session.-+---@@ -35,7 +35,7 @@ Open your browser and go to:Or click **“Build”** in the navigation bar.-+---@@ -43,12 +43,12 @@ Or click **“Build”** in the navigation bar.1. Click the **"Blocks"** button on the left sidebar.-+2. In the search bar, type `AI Text Generator`.3. Drag the block into the canvas.-+---@@ -56,19 +56,15 @@ Or click **“Build”** in the navigation bar.Click the AI Text Generator block to configure it.-In the **LLM Model** dropdown, select one of the supported models from AI/ML API:+In the **LLM Model** dropdown, select the model currently available from AI/ML API:-+-| Model ID | Speed | Reasoning Quality | Best For |-| ---------------------------------------------- | ------ | ----------------- | ------------------------ |-| `Qwen/Qwen2.5-72B-Instruct-Turbo` | Medium | High | Text-based tasks |-| `nvidia/llama-3.1-nemotron-70b-instruct` | Medium | High | Analytics and reasoning |-| `meta-llama/Llama-3.3-70B-Instruct-Turbo` | Low | Very High | Complex multi-step tasks |-| `meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo` | Low | Very High | Deep reasoning |-| `meta-llama/Llama-3.2-3B-Instruct-Turbo` | High | Medium | Fast responses |+| Model | Model ID |+| --- | --- |+| Llama 3.3 70B Instruct Turbo | `meta-llama/Llama-3.3-70B-Instruct-Turbo` |-> ✅ These models are available via OpenAI-compatible API from [AI/ML API](https://aimlapi.com/app/?utm_source=autogpt&utm_medium=github&utm_campaign=integration)+> ✅ This model is available via the OpenAI-compatible API from [AI/ML API](https://aimlapi.com/app/?utm_source=autogpt&utm_medium=github&utm_campaign=integration)---@@ -82,13 +78,13 @@ Inside the **AI Text Generator** block:🔐 You can get your key from:[https://aimlapi.com/app/keys/](https://aimlapi.com/app/keys?utm_source=autogpt&utm_medium=github&utm_campaign=integration)-+-+-+-+---@@ -99,7 +95,7 @@ Click the **“Save”** button at the top-right of the builder interface:1. Give your agent a name (e.g., `aimlapi_test_agent`).2. Click **“Save Agent”** to confirm.-+> 💡 Saving allows reuse, scheduling, and chaining in larger workflows.@@ -112,7 +108,7 @@ From the workspace:1. Press **“Run”** next to your saved agent.2. The request will be sent to the selected AI/ML API model.-+---@@ -122,7 +118,7 @@ From the workspace:2. Check the **Output** panel below it.3. You can copy, export, or pass the result to further blocks.-+---diff --git a/docs/platform/codex-subscription-preview.md b/docs/platform/codex-subscription-preview.mdindex e7fcea7701..4b1dea7139 100644--- a/docs/platform/codex-subscription-preview.md+++ b/docs/platform/codex-subscription-preview.md@@ -36,9 +36,9 @@ flag environment variables. The stack mounts `/run/autogpt-codex` as amemory-backed temporary filesystem for REST, executor, and Copilot executorprocesses.-Open `http://localhost:3000`, then open **Settings > Integrations** and connect-**Codex**. In the sign-in window, open the ChatGPT verification page and enter-the one-time code.+Open `http://localhost:3000`, then open **Settings > Integrations**, choose+**OpenAI**, select **ChatGPT**, and click **Sign in with ChatGPT**. In the sign-in+window, open the ChatGPT verification page and enter the one-time code.### Test the Code Generation block@@ -59,9 +59,9 @@ independently of public API model names.automatically and no connection selector is shown. If multiple subscriptiontransports are connected, choose the desired connection before sending thefirst message.-3. Choose Fast or Thinking and Balanced or Advanced as usual, then send the- message. Text and permitted AutoGPT tool calls stream through the existing- AutoPilot event surface.+3. Choose Balanced or Advanced as available, then send the message. Text and+ permitted AutoGPT tool calls stream through the existing AutoPilot event+ surface.The route and credential are stored on the new session and are immutable forthat session. Start another task to use a different connected subscription@@ -70,15 +70,20 @@ queued turn; queued messages carry only the credential ID, never the OAuthtokens. A missing, revoked, or busy Codex credential fails visibly. There is nosilent fallback to an AutoGPT-funded model or another user's credential.-The mode and model controls select a Codex route from the shared model catalog,-then validate it against the models advertised by the connected account. The-preview maps Fast/Balanced to GPT-5.6 Luna, Fast/Advanced and Thinking/Balanced-to GPT-5.6 Terra, and Thinking/Advanced to GPT-5.6 Sol when the account exposes-them. It otherwise uses the visible account default. File attachments,-agent-building tools, and SDK sub-sessions use the same Claude Agent SDK path as-platform-funded AutoPilot. Builder-panel-bound sessions remain platform-funded-in this preview because their persistent session is created without an AI-connection selector.+The visible model tier and the server-selected execution path select a Codex+route from the shared model catalog, then validate it against the models+advertised by the connected account.++| Execution path | Balanced | Advanced |+| --- | --- | --- |+| Fast | GPT-5.6 Luna | GPT-5.6 Terra |+| Claude Agent SDK | GPT-5.6 Terra | GPT-5.6 Sol |++If the account does not expose the mapped model, the transport uses the visible+account default. File attachments, agent-building tools, and SDK sub-sessions+use the same Claude Agent SDK path as platform-funded AutoPilot.+Builder-panel-bound sessions remain platform-funded in this preview because+their persistent session is created without an AI connection selector.The Copilot executor keeps one exclusive credential lease and one Codex runtimeper connected account, then multiplexes overlapping chats onto separate Codexdiff --git a/docs/platform/gemini.md b/docs/platform/gemini.mdindex 77b43e0e67..5c369b7720 100644--- a/docs/platform/gemini.md+++ b/docs/platform/gemini.md@@ -51,14 +51,16 @@ Click the AI Text Generator block to configure it.In the **LLM Model** dropdown, select one of the available Gemini models:-| Model | Description | Best For |-|-------|-------------|----------|-| `google/gemini-3-pro-preview` | Latest Gemini model | Complex reasoning, coding, multimodal tasks |-| `google/gemini-2.5-pro-preview-03-25` | High capability model | Complex reasoning, coding, multimodal tasks |-| `google/gemini-2.5-flash` | Fast, efficient performance | Quick responses, high-volume tasks |-| `google/gemini-2.5-flash-lite-preview-06-17` | Lightweight preview | Simple tasks, low latency requirements |-| `google/gemini-2.0-flash-001` | Balanced speed and quality | General-purpose applications |-| `google/gemini-2.0-flash-lite-001` | Lightweight, cost-effective | Simple tasks, low latency requirements |+| Model | Model ID |+| --- | --- |+| Gemini 3.1 Pro Preview | `google/gemini-3.1-pro-preview` |+| Gemini 3.1 Flash Lite Preview | `google/gemini-3.1-flash-lite-preview` |+| Gemini 3 Flash Preview | `google/gemini-3-flash-preview` |+| Gemini 2.5 Pro | `google/gemini-2.5-pro` |+| Gemini 2.5 Flash | `google/gemini-2.5-flash` |+| Gemini 2.5 Flash Lite | `google/gemini-2.5-flash-lite` |+| Gemini 2.0 Flash 001 | `google/gemini-2.0-flash-001` |+| Gemini 2.0 Flash Lite 001 | `google/gemini-2.0-flash-lite-001` |> Select the models prefixed with `google/` in the dropdown.@@ -144,7 +146,7 @@ Pricing varies by model tier and usage volume.### Rate Limiting- Free tier has request limits per minute- Upgrade to paid tier for production usage-- Consider using `google/gemini-2.0-flash-lite` for cost-effective high-volume tasks+- Consider using `google/gemini-2.5-flash-lite` for lower-cost, high-volume tasks### Context Length Errors- Each Gemini model has a maximum context windowdiff --git a/docs/platform/integrating/api-guide.md b/docs/platform/integrating/api-guide.mdindex 19d210af91..09b81f6638 100644--- a/docs/platform/integrating/api-guide.md+++ b/docs/platform/integrating/api-guide.md@@ -37,7 +37,7 @@ OAuth is ideal for:- "Sign in with AutoGPT" (SSO, Single Sign-On) functionality- Applications that need user-specific permissions-See the [SSO Integration Guide](sso-guide.md) for complete OAuth implementation details.+See the [OAuth Integration Guide](oauth-guide.md) for complete OAuth implementation details.## Available Scopes@@ -48,9 +48,11 @@ When using OAuth, request only the scopes your application needs:| `IDENTITY` | Read user ID, e-mail, and timezone || `EXECUTE_GRAPH` | Run agents || `READ_GRAPH` | Read agent run results |+| `WRITE_GRAPH` | Create and update agent graphs || `EXECUTE_BLOCK` | Run individual blocks || `READ_BLOCK` | Read block definitions || `READ_STORE` | Access the agent store |+| `WRITE_LIBRARY` | Add agents to your library || `USE_TOOLS` | Use platform tools || `MANAGE_INTEGRATIONS` | Create and update user integrations || `READ_INTEGRATIONS` | Read user integration status |@@ -69,7 +71,7 @@ curl -H "X-API-Key: YOUR_API_KEY" \### Using OAuth1. Register an OAuth application (contact platform administrator)-2. Implement the OAuth flow as described in the [SSO Guide](sso-guide.md)+2. Implement the OAuth flow as described in the [OAuth Guide](oauth-guide.md)3. Use the obtained access token:```bashdiff --git a/docs/platform/ollama.md b/docs/platform/ollama.mdindex 46c02ec93e..41b0ae9e2b 100644--- a/docs/platform/ollama.md+++ b/docs/platform/ollama.md@@ -19,10 +19,10 @@ To properly set up Ollama for network access, choose one of these methods:1. Open the Ollama desktop application2. Go to **Settings** and toggle **"Expose Ollama to the network"**- + 3. Click on the model name field in the "New Chat" window4. Search for "llama3.2" (or your preferred model)- + 5. Click on it to start the download and load the model to be used??? note "Method B: Using Docker (Alternative)"@@ -99,15 +99,16 @@ This command starts both the backend and frontend services. Once running, visitNow that both Ollama and the AutoGPT platform are running, we can use Ollama with AutoGPT:1. Add an AI Text Generator block to your workspace (it can work with any AI LLM block but for this example will be using the AI Text Generator block):- + 2. **Configure the API Key field**: Enter any value (e.g., "dummy" or "not-needed") since Ollama doesn't require authentication.3. In the "LLM Model" dropdown, select "llama3.2" (This is the model we downloaded earlier)- + > **Compatible Models**: The following Ollama models are available in AutoGPT by default:> - `llama3.2` (Recommended for most use cases)+ > - `llama3.3`> - `llama3`> - `llama3.1:405b`> - `dolphin-mistral:latest`@@ -137,15 +138,15 @@ Now that both Ollama and the AutoGPT platform are running, we can use Ollama wit192.168.0.39:11434```- + > **Important**: Since AutoGPT runs in Docker containers, you must use your host machine's IP address instead of `localhost` or `127.0.0.1`. Docker containers cannot reach `localhost` on the host machine.5. Add prompts to your AI block, save the graph, and run it:- + That's it! You've successfully setup the AutoGPT platform and made a LLM call to Ollama.-+### Using Ollama on a Remote Server with AutoGPT@@ -174,7 +175,7 @@ Then you can use the same steps above but you need to add the Ollama server's IP192.168.0.39:11434```-+## Add Custom Models (Advanced)