Connect AI Providers
QCoder supports 11 AI providers out of the box. Add your API key and start chatting in seconds.
Supported Providers
QCoder works with any OpenAI-compatible API, plus native integrations for major providers. Here is the full list:
| Provider | API Key Format | Context Window | Tool Calling |
|---|---|---|---|
| --- | --- | --- | --- |
| OpenAI | sk-... | 128K tokens | Native |
| Anthropic | sk-ant-... | 200K tokens | Native |
| DeepSeek | sk-... | 64K tokens | Native |
| Google Gemini | AIza... | 1M+ tokens | Native |
| Groq | gsk_... | 64K tokens | Native |
| Ollama | None (local) | 32K tokens | Via prompting |
| OpenRouter | sk-or-... | Varies by model | Varies |
| xAI / Grok | xai-... | Varies | Native |
| Mistral | API key | 128K tokens | Native |
| LM Studio | None (local) | Varies | Via prompting |
| Custom HTTP | Configurable | Configurable | Configurable |
Adding a Provider
- Open the chat panel and click the gear icon to open Settings.
- Select the API Config tab.
- Choose your provider from the Provider dropdown.
- Paste your API key into the API Key field.
- (Optional) Override the Base URL if you are using a proxy or self-hosted endpoint.
- Select a Model from the dropdown. Models are fetched dynamically from the provider when possible.
- Click Save or simply close the Settings panel -- changes are saved automatically.
You can configure multiple providers and switch between them at any time using the model selector in the chat header.
Local Providers (Free)
Ollama and LM Studio let you run models entirely on your machine with no API key and no cost.
Ollama setup:
1. Install Ollama from [ollama.com](https://ollama.com).
2. Pull a model: ollama pull llama3.1 or ollama pull codellama.
3. In QCoder Settings > API Config, select Ollama as the provider.
4. The base URL defaults to http://localhost:11434. Change it only if you run Ollama on a different port.
5. Select your pulled model from the dropdown and start chatting.
LM Studio setup: 1. Install LM Studio from [lmstudio.ai](https://lmstudio.ai). 2. Download a model through the LM Studio interface. 3. Start the local server in LM Studio (it will show you the port). 4. In QCoder, select LM Studio as the provider and enter the server URL.
OpenRouter (Multi-Provider Proxy)
OpenRouter acts as a single API gateway to hundreds of models from multiple providers. This is useful if you want access to many models with one API key.
- Sign up at [openrouter.ai](https://openrouter.ai) and generate an API key.
- In QCoder Settings, select OpenRouter as the provider.
- Paste your
sk-or-...key. - Browse and select from the full model catalog.
OpenRouter supports pay-per-use pricing and routes requests to the cheapest available endpoint for each model.
Rate Limiting and Retries
QCoder includes built-in retry logic for rate-limited or temporarily failed API calls:
| Setting | Value |
|---|---|
| --- | --- |
| Max retries | 3 |
| Base delay | 2,000 ms |
| Max delay | 30,000 ms |
| Min interval between requests | 1,000 ms |
Retries use exponential backoff. If a provider returns a rate-limit error (HTTP 429), QCoder waits and retries automatically. You do not need to do anything -- just wait for the response to arrive.
QCoder also includes a temperature auto-retry feature. If a model does not support the temperature parameter (such as OpenAI's o1 and o3 reasoning models), QCoder detects the error and automatically retries the request without temperature.