> ## Documentation Index
> Fetch the complete documentation index at: https://mintlify.com/skydiscover-ai/skydiscover/llms.txt
> Use this file to discover all available pages before exploring further.

# Model Providers

> Configure OpenAI, Gemini, Anthropic, DeepSeek, and local models with SkyDiscover

## Overview

SkyDiscover supports any OpenAI-compatible API endpoint. Provider-specific configurations are handled automatically based on model name prefixes.

## Supported Providers

<CardGroup cols={2}>
  <Card title="OpenAI" icon="openai">
    GPT-4, GPT-5, o1, o3, o4 models
  </Card>

  <Card title="Google Gemini" icon="google">
    Gemini 2.0, 3.0 models
  </Card>

  <Card title="Anthropic" icon="brain">
    Claude 3.5, 3.7, 4.0 models
  </Card>

  <Card title="DeepSeek" icon="code">
    DeepSeek-V3, DeepSeek-Coder
  </Card>

  <Card title="Mistral" icon="wind">
    Mistral Large, Codestral
  </Card>

  <Card title="Local Models" icon="server">
    Ollama, vLLM, any OpenAI-compatible
  </Card>
</CardGroup>

## OpenAI

### Setup

```bash theme={null}
export OPENAI_API_KEY="sk-..."
```

### Configuration

<CodeGroup>
  ```yaml Config File theme={null}
  llm:
    models:
      - name: "gpt-5"
        weight: 1.0
      - name: "o1"
        reasoning_effort: "high"
        weight: 0.3
    api_base: "https://api.openai.com/v1"
    temperature: 0.7
    max_tokens: 32000
  ```

  ```bash CLI theme={null}
  skydiscover-run init.py eval.py -m gpt-5
  ```

  ```python Python API theme={null}
  from skydiscover import run_discovery

  result = run_discovery(
      evaluator="eval.py",
      initial_program="init.py",
      model="gpt-5",
      iterations=50,
  )
  ```
</CodeGroup>

### Supported Models

* `gpt-5` (latest GPT-5)
* `gpt-4o`, `gpt-4o-mini`
* `o1`, `o1-mini`, `o1-pro`
* `o3`, `o3-mini`
* `o4`, `o4-mini`

### Reasoning Models (o1/o3/o4)

For o-series models, set reasoning effort:

```yaml theme={null}
llm:
  models:
    - name: "o1"
      reasoning_effort: "high"  # "low", "medium", "high"
      weight: 1.0
  temperature: 1.0  # o-series uses fixed temperature
```

## Google Gemini

### Setup

```bash theme={null}
export GEMINI_API_KEY="AIza..."
# or
export GOOGLE_API_KEY="AIza..."
```

### Configuration

<CodeGroup>
  ```yaml Config File theme={null}
  llm:
    models:
      - name: "gemini/gemini-3-pro"
        weight: 1.0
    api_base: "https://generativelanguage.googleapis.com/v1beta/openai/"
    temperature: 0.7
    max_tokens: 32000
  ```

  ```bash CLI theme={null}
  skydiscover-run init.py eval.py -m gemini/gemini-3-pro
  ```

  ```python Python API theme={null}
  result = run_discovery(
      evaluator="eval.py",
      model="gemini/gemini-3-pro",
      iterations=50,
  )
  ```
</CodeGroup>

### Supported Models

* `gemini/gemini-3-pro` (Gemini 3.0 Pro)
* `gemini/gemini-3-flash` (Gemini 3.0 Flash)
* `gemini/gemini-2.0-flash-exp`

<Note>
  The `gemini/` prefix is required for auto-detection. SkyDiscover strips it before sending to the API.
</Note>

## Anthropic Claude

### Setup

```bash theme={null}
export ANTHROPIC_API_KEY="sk-ant-..."
```

### Configuration

<CodeGroup>
  ```yaml Config File theme={null}
  llm:
    models:
      - name: "claude-3-7-sonnet"
        weight: 1.0
    api_base: "https://api.anthropic.com/v1/"
    temperature: 0.7
    max_tokens: 32000
  ```

  ```bash CLI theme={null}
  skydiscover-run init.py eval.py -m claude-3-7-sonnet
  ```

  ```python Python API theme={null}
  result = run_discovery(
      evaluator="eval.py",
      model="claude-3-7-sonnet",
      iterations=50,
  )
  ```
</CodeGroup>

### Supported Models

* `claude-3-7-sonnet`
* `claude-3-5-sonnet`
* `claude-4-opus` (when available)

## DeepSeek

### Setup

```bash theme={null}
export DEEPSEEK_API_KEY="sk-..."
```

### Configuration

<CodeGroup>
  ```yaml Config File theme={null}
  llm:
    models:
      - name: "deepseek-chat"
        weight: 1.0
    api_base: "https://api.deepseek.com/v1"
    temperature: 0.7
    max_tokens: 32000
  ```

  ```bash CLI theme={null}
  skydiscover-run init.py eval.py -m deepseek-chat
  ```
</CodeGroup>

### Supported Models

* `deepseek-chat` (DeepSeek-V3)
* `deepseek-coder`

## Mistral

### Setup

```bash theme={null}
export MISTRAL_API_KEY="..."
```

### Configuration

```yaml theme={null}
llm:
  models:
    - name: "mistral-large"
      weight: 1.0
  api_base: "https://api.mistral.ai/v1"
```

## Azure OpenAI

### Setup

```bash theme={null}
export AZURE_API_KEY="..."
export AZURE_API_BASE="https://your-resource.openai.azure.com"
```

### Configuration

```yaml theme={null}
llm:
  models:
    - name: "gpt-5"  # Your deployment name
      api_base: "https://your-resource.openai.azure.com"
      api_key: ${AZURE_API_KEY}
      weight: 1.0
```

## Local Models

### Ollama

<Steps>
  <Step title="Install Ollama">
    ```bash theme={null}
    curl -fsSL https://ollama.com/install.sh | sh
    ```
  </Step>

  <Step title="Pull a Model">
    ```bash theme={null}
    ollama pull qwen2.5-coder:32b
    ```
  </Step>

  <Step title="Configure SkyDiscover">
    ```yaml theme={null}
    llm:
      models:
        - name: "qwen2.5-coder:32b"
          weight: 1.0
      api_base: "http://localhost:11434/v1"
      api_key: "ollama"  # Dummy key
    ```

    Or via CLI:

    ```bash theme={null}
    skydiscover-run init.py eval.py \
      --api-base http://localhost:11434/v1 \
      -m qwen2.5-coder:32b
    ```
  </Step>
</Steps>

### vLLM

<Steps>
  <Step title="Install vLLM">
    ```bash theme={null}
    pip install vllm
    ```
  </Step>

  <Step title="Start vLLM Server">
    ```bash theme={null}
    vllm serve deepseek-ai/DeepSeek-Coder-V2-Instruct \
      --host 0.0.0.0 \
      --port 8000 \
      --tensor-parallel-size 4
    ```
  </Step>

  <Step title="Configure SkyDiscover">
    ```yaml theme={null}
    llm:
      models:
        - name: "deepseek-ai/DeepSeek-Coder-V2-Instruct"
          weight: 1.0
      api_base: "http://localhost:8000/v1"
      api_key: "EMPTY"  # vLLM doesn't require auth by default
    ```
  </Step>
</Steps>

### LM Studio

<Steps>
  <Step title="Install LM Studio">
    Download from [lmstudio.ai](https://lmstudio.ai/)
  </Step>

  <Step title="Load a Model">
    Search and download `Qwen/Qwen2.5-Coder-32B-Instruct-GGUF`
  </Step>

  <Step title="Start Local Server">
    In LM Studio: `Local Server` → `Start Server` (default port 1234)
  </Step>

  <Step title="Configure SkyDiscover">
    ```bash theme={null}
    skydiscover-run init.py eval.py \
      --api-base http://localhost:1234/v1 \
      -m qwen2.5-coder-32b
    ```
  </Step>
</Steps>

## Multiple Providers

Mix and match providers in a single run:

```yaml theme={null}
llm:
  models:
    - name: "gpt-5"
      weight: 0.4
      api_base: "https://api.openai.com/v1"
      api_key: ${OPENAI_API_KEY}
    
    - name: "gemini/gemini-3-pro"
      weight: 0.3
      api_base: "https://generativelanguage.googleapis.com/v1beta/openai/"
      api_key: ${GEMINI_API_KEY}
    
    - name: "claude-3-7-sonnet"
      weight: 0.2
      api_base: "https://api.anthropic.com/v1/"
      api_key: ${ANTHROPIC_API_KEY}
    
    - name: "qwen2.5-coder:32b"
      weight: 0.1
      api_base: "http://localhost:11434/v1"
      api_key: "ollama"
```

SkyDiscover samples models proportionally to their weights.

## Model-Specific Settings

### Per-Model Parameters

```yaml theme={null}
llm:
  models:
    - name: "gpt-5"
      weight: 0.5
      temperature: 0.8      # Higher creativity
      max_tokens: 40000
    
    - name: "o1"
      weight: 0.5
      temperature: 1.0      # Fixed for o-series
      reasoning_effort: "high"
      max_tokens: 32000
  
  # Defaults for unspecified parameters
  temperature: 0.7
  max_tokens: 32000
  timeout: 600
```

### Custom Client Initialization

For advanced use cases, pass a custom LLM client:

```python theme={null}
from skydiscover import run_discovery
from skydiscover.config import Config, LLMModelConfig
from openai import AsyncOpenAI

def init_custom_client():
    return AsyncOpenAI(
        base_url="http://localhost:8000/v1",
        api_key="custom-key",
        timeout=600,
    )

config = Config()
config.llm.models = [
    LLMModelConfig(
        name="custom-model",
        init_client=init_custom_client,
        weight=1.0,
    )
]

result = run_discovery(
    evaluator="eval.py",
    config=config,
    iterations=50,
)
```

## Provider Auto-Detection

SkyDiscover automatically detects providers from model names:

| Model Name Pattern           | Provider  | API Base                                                   |
| ---------------------------- | --------- | ---------------------------------------------------------- |
| `gpt-*`, `o1*`, `o3*`, `o4*` | OpenAI    | `https://api.openai.com/v1`                                |
| `gemini-*`, `gemini/*`       | Gemini    | `https://generativelanguage.googleapis.com/v1beta/openai/` |
| `claude-*`                   | Anthropic | `https://api.anthropic.com/v1/`                            |
| `deepseek-*`                 | DeepSeek  | `https://api.deepseek.com/v1`                              |
| `mistral-*`                  | Mistral   | `https://api.mistral.ai/v1`                                |
| `command-*`                  | Cohere    | `https://api.cohere.com/v1`                                |
| Other                        | OpenAI    | `https://api.openai.com/v1`                                |

### Override Detection

Use `provider/model` syntax or explicit `api_base`:

```yaml theme={null}
llm:
  models:
    # Explicit provider prefix
    - name: "gemini/gemini-3-pro"
      weight: 1.0
    
    # Explicit api_base override
    - name: "my-custom-model"
      api_base: "http://localhost:8000/v1"
      weight: 1.0
```

## API Key Priority

SkyDiscover resolves API keys in this order:

1. **Per-model config:**
   ```yaml theme={null}
   llm:
     models:
       - name: "gpt-5"
         api_key: "sk-..."
   ```

2. **Provider environment variable:**
   ```bash theme={null}
   export GEMINI_API_KEY="AIza..."
   export ANTHROPIC_API_KEY="sk-ant-..."
   export DEEPSEEK_API_KEY="sk-..."
   ```

3. **Generic OPENAI\_API\_KEY:**
   ```bash theme={null}
   export OPENAI_API_KEY="sk-..."
   ```

<Tip>
  Always set provider-specific keys when using multiple providers to avoid auth errors.
</Tip>

## Troubleshooting

<AccordionGroup>
  <Accordion title="Authentication Failed">
    **Error:** `401 Unauthorized`

    **Fix:**

    1. Check your API key is correct
    2. Verify the key is exported: `echo $OPENAI_API_KEY`
    3. Ensure provider-specific key is set (e.g., `GEMINI_API_KEY` for Gemini)
    4. Try hardcoding the key in config temporarily to isolate the issue
  </Accordion>

  <Accordion title="Connection Refused (Local Models)">
    **Error:** `Connection refused` or `Failed to connect`

    **Fix:**

    1. Check server is running: `curl http://localhost:11434/v1/models`
    2. Verify port matches config
    3. For Ollama: `ollama serve` should show "Listening on 127.0.0.1:11434"
    4. For vLLM: Check logs for startup errors
  </Accordion>

  <Accordion title="Model Not Found">
    **Error:** `The model 'xyz' does not exist`

    **Fix:**

    1. For Ollama: `ollama list` to see available models
    2. For cloud providers: Check spelling and model availability
    3. Verify provider prefix: `gemini/gemini-3-pro` not just `gemini-3-pro`
  </Accordion>

  <Accordion title="Rate Limit Exceeded">
    **Error:** `429 Too Many Requests`

    **Fix:**

    1. Reduce `max_parallel_iterations` in config
    2. Add retry delay: `retry_delay: 10`
    3. Use multiple API keys with separate model entries
    4. Switch to local models for unlimited requests
  </Accordion>
</AccordionGroup>

## Best Practices

<CardGroup cols={2}>
  <Card title="Use Environment Variables" icon="shield">
    Never commit API keys. Use env vars or `.env` files.

    ```bash theme={null}
    # .env
    OPENAI_API_KEY=sk-...
    GEMINI_API_KEY=AIza...
    ```
  </Card>

  <Card title="Mix Local and Cloud" icon="cloud">
    Use cheap local models for exploration, cloud models for exploitation.

    ```yaml theme={null}
    models:
      - name: "qwen2.5-coder:32b"
        weight: 0.7  # Exploration
      - name: "gpt-5"
        weight: 0.3  # Exploitation
    ```
  </Card>

  <Card title="Set Timeouts" icon="clock">
    Increase timeout for large models or slow servers.

    ```yaml theme={null}
    llm:
      timeout: 1200  # 20 minutes
    ```
  </Card>

  <Card title="Monitor Costs" icon="dollar-sign">
    Use `gpt-4o-mini` or local models during development.

    ```yaml theme={null}
    models:
      - name: "gpt-4o-mini"  # $0.15/M tokens
        weight: 1.0
    ```
  </Card>
</CardGroup>

## Next Steps

<CardGroup cols={2}>
  <Card title="Configuration" icon="sliders" href="/guides/configuration">
    Full LLM configuration reference
  </Card>

  <Card title="Running Discovery" icon="rocket" href="/guides/running-discovery">
    Start your first discovery run
  </Card>
</CardGroup>