Skip to main content

Connecting an AI Model Provider

Overview​

AI Model Providers are companies that develop and host large language models (LLMs) like Claude, GPT, and Gemini. To use PostQode, you need to connect to an AI Model Provider by obtaining an API key from them. This API key allows PostQode to securely communicate with the provider's AI models on your behalf.

What is an AI Model Provider?

An AI Model Provider is a company that:

  • Develops and trains large language models (LLMs)
  • Hosts these models on their infrastructure
  • Provides API access to their models through API keys
  • Charges based on usage (tokens processed)

Examples include Anthropic (Claude), OpenAI (GPT), Google (Gemini), and aggregators like OpenRouter that provide access to multiple models.

PostQode requires an API key from an AI Model Provider to function. The API key authenticates your requests and allows PostQode to access the AI models.

We recommend these options for accessing the latest and most powerful AI models:

Latest Models Available (2026)​

  • Claude Sonnet 5, Claude Opus 4.8 & Claude 3.7 Sonnet - Anthropic's flagship agentic coding and deep reasoning models with adaptive thinking and industry-leading diff precision.
  • GPT-5.5, GPT-5.2 & o3-mini / o4-mini - OpenAI's frontier agentic coding and reasoning models for complex multi-step tasks.
  • Google Gemini 3.7 Flash & Gemini 3.1 Pro - High-speed, 1M+ token context multimodal models for codebase-wide understanding and rapid iteration.
  • DeepSeek R1 & V3 - State-of-the-art open-weights reasoning (deepseek-reasoner) and coding (deepseek-chat) models.
  • Qwen 2.5 Coder & Meta Llama 3.3 - High-performance open models for cost-effective local and cloud execution.

Provider Options​

  • PostQode Managed Provider (Recommended for Instant Access): Provides instant access to frontier and open models with competitive per-token pricing and free trial credits. Includes popular models:

    • Claude Sonnet 5 (anthropic/claude-sonnet-5) - Flagship coding model with optimal speed, cost, and diff quality (Recommended)
    • OpenAI GPT-5.5 & GPT-4o (openai/gpt-5.5, openai/gpt-4o) - High-depth multi-step logic and execution
    • Google Gemini 3.7 Flash & 2.5 Flash (google/gemini-3.7-flash) - 1M token context for repository-wide exploration (Free tier available)
    • DeepSeek R1 & V3 (deepseek/deepseek-r1, deepseek/deepseek-chat) - Advanced reasoning and code generation
    • Qwen 2.5 Coder 32B / 72B - High-performance coding model
    • Meta Llama 3.3 70B - Powerful open-source foundation model

  • Anthropic: Direct access to Claude models (claude-sonnet-5, claude-fable-5, claude-opus-4-8, claude-sonnet-4-6, claude-haiku-4-5, and legacy claude-3-7-sonnet) using your API key from the Anthropic Console.

  • OpenAI: Direct access to gpt-5.5, gpt-5.2, gpt-4o, gpt-4o-mini, o3-mini, o4-mini, and o1 via your API key from the OpenAI Platform.

  • Google Gemini: Direct access to gemini-3.7-flash, gemini-3.1-pro-preview, gemini-2.5-pro, and gemini-2.5-flash via your API key from Google AI Studio.

  • DeepSeek: Direct access to deepseek-chat and deepseek-reasoner (R1) via your API key from the DeepSeek Platform.

  • OpenRouter & LLM Routers: Aggregated access to 200+ models across all providers using a single API key from OpenRouter.

  • OpenAI Compatible & Local Servers: Connect Ollama, LM Studio, vLLM, LiteLLM, or self-hosted gateway endpoints. Requires three primary settings: Base URL, API Key, and typing your custom Model ID.

Using the PostQode Provider​

By default when you install the PostQode extension, you'll be prompted to sign in or create an account with PostQode at https://app.postqode.ai.

This will walk you through the account setup and automatically configure PostQode to get you started. If you'd rather use another AI Model Provider (BYOK), you can configure your API key as described below.

Using Another AI Model Provider (BYOK)​

Getting Your API Key​

Option 1: LLM Routers (OpenRouter)​

  1. Go to openrouter.ai.
  2. Sign in with your account.
  3. Navigate to API keys and generate a new key.
  4. Copy your API key for PostQode.

Option 2: Direct AI Model Providers​

Anthropic​
  1. Go to the Anthropic Console.
  2. Sign up or log in.
  3. Navigate to Settings ➔ API Keys and click Create Key.
  4. Copy your key (sk-ant-...).
OpenAI​
  1. Go to the OpenAI Platform.
  2. Sign up or log in.
  3. Navigate to API Keys ➔ Create new secret key.
  4. Copy your key (sk-...).
Google Gemini​
  1. Go to Google AI Studio.
  2. Sign up or log in with your Google account.
  3. Navigate to Get API key and click Create API key.
  4. Copy your key (AIza...).
DeepSeek​
  1. Go to the DeepSeek Platform.
  2. Sign up or log in.
  3. Navigate to API Keys ➔ API Keys and click Create new API key.
  4. Copy your key (sk-...).

Configuring the AI Model Provider in PostQode​

Once you have your API key from your chosen provider:

  1. Open the PostQode sidebar by clicking the PostQode icon () in the Side Bar.
  2. Click the Settings gear icon in the top-right corner.
  3. In the Provider dropdown, select your desired provider (e.g. PostQode, Anthropic, OpenAI, Google Gemini, DeepSeek, OpenRouter, Ollama, OpenAI Compatible).
  4. Paste your API key into the API Key field.
  5. Select or enter your model:
    • For PostQode Managed Provider: Choose anthropic/claude-sonnet-5 (Recommended), google/gemini-3.7-flash, or openai/gpt-5.5 from the curated list.
    • For Anthropic: Choose claude-sonnet-5 (Recommended) or claude-opus-4-8 from the dropdown.
    • For OpenAI: Choose gpt-5.5, gpt-4o, or o3-mini from the dropdown.
    • For Google Gemini: Choose gemini-3.7-flash (Recommended) or gemini-2.5-pro from the dropdown.
    • For DeepSeek: Choose deepseek-chat or deepseek-reasoner from the dropdown.
    • For OpenRouter: Choose your desired model ID (e.g. anthropic/claude-sonnet-5, deepseek/deepseek-r1:free).
    • For OpenAI Compatible / Local Servers: Enter your Base URL (e.g. http://localhost:11434/v1), your API Key, and manually type the target Model ID (e.g. qwen2.5-coder:32b).
  6. Click Save Settings to start using PostQode.

Next Steps​

After connecting your AI Model Provider, you're ready to start using PostQode! Check out Your First Task to learn how to give PostQode its first assignment.

Is this documentation incorrect or incomplete? Report an issue on GitHub