Guide

Huddle01 AI Inference gives you one OpenAI-compatible endpoint for multiple model providers.

What is inference?

Inference means sending input (messages, prompt, or multimodal data) to a model and getting generated output in response.

With Huddle01:

• You use one API key.
• You hit one base URL.
• You can switch models by changing only the model field.

Endpoint and auth

• Base URL: https://gru.huddle01.io/v1
• Auth header: Authorization: Bearer <HUDDLE_API_KEY>
• Compatibility: OpenAI-compatible SDKs and HTTP APIs

Quick navigation

Quick start flow

1. Create or copy your Huddle01 API key from dashboard.
2. Set your client base_url to https://gru.huddle01.io/v1.
3. Call chat/completions with one of the supported model IDs.
4. Read usage and spend from your existing dashboard/billing views.

Apps and IDEs

Most apps that support OpenAI-compatible providers need the same three fields:

OpenCode

Create or update opencode.json in your project root, or place it in your global OpenCode config directory.

{
  "$schema": "https://opencode.ai/config.json",
  "provider": {
    "huddle01": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "Huddle01 AI Inference",
      "options": {
        "baseURL": "https://gru.huddle01.io/v1",
        "apiKey": "{env:HUDDLE_API_KEY}"
      },
      "models": {
        "qwen3-coder": {
          "name": "Qwen3 Coder"
        },
        "deepseek-v3.2": {
          "name": "DeepSeek V3.2"
        },
        "glm-4.7": {
          "name": "GLM 4.7"
        },
        "minimax-m2.5": {
          "name": "MiniMax M2.5"
        }
      }
    }
  }
}

Then set your key and choose a model in OpenCode:

export HUDDLE_API_KEY="your_huddle01_key"
opencode

Inside OpenCode, run /models and select one of the huddle01 models.

Other apps

For tools like Cursor extensions, Continue, Cline, Open WebUI, LangChain, or custom OpenAI SDK clients, use the same setup:

1. Choose OpenAI Compatible as the provider.
2. Paste https://gru.huddle01.io/v1 as the base URL.
3. Paste your Huddle01 API key.
4. Set the model ID, for example qwen3-coder, deepseek-v3.2, glm-4.7, or minimax-m2.5.

If the app asks for a full endpoint instead of a base URL, use https://gru.huddle01.io/v1/chat/completions.

Request lifecycle

1. Your app sends a request to POST /chat/completions.
2. Huddle01 authenticates the API key and validates request shape.
3. The AI Inference service routes your request to the selected model/provider.
4. Response is normalized into OpenAI-compatible format.
5. You receive generated output and usage metadata.

What you can build

• AI chat and assistant experiences
• Content generation and summarization
• Code generation and developer tools
• Workflow automations with structured prompting
• Multi-model routing with one integration surface

Compatibility

• Works with OpenAI-compatible SDKs
• Works with direct HTTP calls (curl, backend services, serverless functions)
• Supports model switching without client rewrites

Best practices

• Keep API keys on the server side only.
• Use environment variables and secret managers, never hardcode keys.
• Start with lower-cost models for non-critical traffic.
• Add retry + timeout handling in production clients.
• Log model, latency, token usage, and request ID for debugging.

Common mistakes to avoid

• Using the wrong base URL (must be https://gru.huddle01.io/v1)
• Sending API key in custom headers instead of Authorization: Bearer ...
• Assuming every model has identical behavior/capabilities
• Not handling transient upstream/network failures

Next steps

1. Go to API Examples for concrete API requests and SDK examples.
2. Check Pricing to pick the right models for your workloads.