> ## 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.

# LLM Configuration

> Configure language models, API settings, and generation parameters for SkyDiscover

## Overview

The `llm` section controls all language model settings including API endpoints, model selection, generation parameters, and retry behavior.

## Basic Configuration

```yaml theme={null}
llm:
  models:
    - name: "gpt-5"
      weight: 1.0
  api_base: "https://api.openai.com/v1"
  temperature: 0.7
  top_p: 0.95
  max_tokens: 32000
  timeout: 600
```

## Model Configuration

### Single Model

Configure a single model for solution generation:

```yaml theme={null}
llm:
  models:
    - name: "gpt-5"
      weight: 1.0
```

### Multiple Models

Use multiple models with weighted sampling:

```yaml theme={null}
llm:
  models:
    - name: "gpt-5"
      weight: 0.7
    - name: "claude-opus-3"
      weight: 0.3
```

SkyDiscover randomly samples models based on their weights. In this example, `gpt-5` is selected 70% of the time.

### Provider-Specific Models

Use the `provider/model` format to specify different providers:

```yaml theme={null}
llm:
  models:
    - name: "openai/gpt-5"
      weight: 0.5
    - name: "anthropic/claude-opus-3"
      weight: 0.3
    - name: "gemini/gemini-3-pro"
      weight: 0.2
```

## Supported Providers

SkyDiscover supports multiple LLM providers with automatic API base URL and key resolution:

<AccordionGroup>
  <Accordion title="OpenAI" icon="openai">
    ```yaml theme={null}
    llm:
      models:
        - name: "gpt-5"
        - name: "gpt-5-mini"
        - name: "o1"
      api_base: "https://api.openai.com/v1"
    ```

    Environment variables: `OPENAI_API_KEY`

    Automatically detected for models starting with: `gpt-`, `o1`, `o3`, `o4`
  </Accordion>

  <Accordion title="Anthropic (Claude)" icon="robot">
    ```yaml theme={null}
    llm:
      models:
        - name: "anthropic/claude-opus-3"
        - name: "anthropic/claude-sonnet-5"
      # api_base auto-configured
    ```

    Environment variables: `ANTHROPIC_API_KEY`

    Automatically detected for models starting with: `claude-`
  </Accordion>

  <Accordion title="Google Gemini" icon="google">
    ```yaml theme={null}
    llm:
      models:
        - name: "gemini/gemini-3-pro"
      api_base: "https://generativelanguage.googleapis.com/v1beta/openai/"
    ```

    Environment variables: `GEMINI_API_KEY`, `GOOGLE_API_KEY`

    Automatically detected for models starting with: `gemini-`
  </Accordion>

  <Accordion title="DeepSeek" icon="brain">
    ```yaml theme={null}
    llm:
      models:
        - name: "deepseek/deepseek-chat"
      api_base: "https://api.deepseek.com/v1"
    ```

    Environment variables: `DEEPSEEK_API_KEY`

    Automatically detected for models starting with: `deepseek-`
  </Accordion>

  <Accordion title="Mistral" icon="wind">
    ```yaml theme={null}
    llm:
      models:
        - name: "mistral/mistral-large"
      api_base: "https://api.mistral.ai/v1"
    ```

    Environment variables: `MISTRAL_API_KEY`

    Automatically detected for models starting with: `mistral-`
  </Accordion>

  <Accordion title="Ollama (Local)" icon="server">
    ```yaml theme={null}
    llm:
      models:
        - name: "ollama/llama3"
          api_base: "http://localhost:11434/v1"
      # No API key required
    ```

    No environment variables required. Specify api\_base for your local Ollama instance.
  </Accordion>

  <Accordion title="vLLM (Local)" icon="server">
    ```yaml theme={null}
    llm:
      models:
        - name: "vllm/custom-model"
          api_base: "http://localhost:8000/v1"
      # No API key required
    ```

    No environment variables required. Specify api\_base for your vLLM server.
  </Accordion>
</AccordionGroup>

## LLMConfig Parameters

Defined in `skydiscover/config.py:151-224`

### API Settings

<ParamField path="api_base" type="str" default="https://api.openai.com/v1">
  Base URL for the LLM API endpoint
</ParamField>

<ParamField path="api_key" type="str" default="None">
  API key for authentication. Auto-resolved from environment if not specified
</ParamField>

### Model Selection

<ParamField path="models" type="List[LLMModelConfig]" default="[]">
  List of models for solution generation. Each model can have individual settings
</ParamField>

<ParamField path="evaluator_models" type="List[LLMModelConfig]" default="[]">
  Models for evaluation. Defaults to same as `models` if not specified
</ParamField>

<ParamField path="guide_models" type="List[LLMModelConfig]" default="[]">
  Models for guide tasks (idea generation, paradigm breakthroughs). Defaults to `models`
</ParamField>

### Generation Parameters

<ParamField path="temperature" type="float" default="0.7">
  Sampling temperature (0.0-2.0). Higher values increase randomness
</ParamField>

<ParamField path="top_p" type="float" default="0.95">
  Nucleus sampling threshold (0.0-1.0). Alternative to temperature
</ParamField>

<ParamField path="max_tokens" type="int" default="32000">
  Maximum tokens to generate in a single completion
</ParamField>

<ParamField path="system_message" type="str" default="system_message">
  System message for the model. Usually set in `prompt` section instead
</ParamField>

### Request Parameters

<ParamField path="timeout" type="int" default="600">
  Request timeout in seconds
</ParamField>

<ParamField path="retries" type="int" default="3">
  Number of retry attempts on failure
</ParamField>

<ParamField path="retry_delay" type="int" default="5">
  Delay in seconds between retries
</ParamField>

### Reasoning Parameters

<ParamField path="reasoning_effort" type="str" default="None">
  Reasoning effort level for o1-series models: `low`, `medium`, or `high`
</ParamField>

## LLMModelConfig Parameters

Per-model configuration (defined in `skydiscover/config.py:121-148`):

<ParamField path="name" type="str" default="None">
  Model identifier (e.g., `gpt-5`, `anthropic/claude-opus-3`)
</ParamField>

<ParamField path="weight" type="float" default="1.0">
  Sampling weight for this model in the pool
</ParamField>

<ParamField path="api_base" type="str" default="None">
  Override API base URL for this specific model
</ParamField>

<ParamField path="api_key" type="str" default="None">
  Override API key for this specific model
</ParamField>

<ParamField path="init_client" type="Callable" default="None">
  Custom LLM client initialization function
</ParamField>

All generation and request parameters can also be overridden per-model.

## Advanced Examples

### LLM-as-a-Judge

Use separate models for generation and evaluation:

```yaml configs/llm_judge.yaml theme={null}
llm:
  # Fast model for generation
  models:
    - name: "gpt-4o-mini"
      weight: 1.0
  
  # Powerful model for evaluation
  evaluator_models:
    - name: "gpt-5"
      weight: 1.0
  
  temperature: 0.7
  max_tokens: 16000
```

### Mixed Provider Setup

```yaml theme={null}
llm:
  models:
    - name: "gpt-5"
      weight: 0.5
      api_base: "https://api.openai.com/v1"
      api_key: ${OPENAI_API_KEY}
    
    - name: "claude-opus-3"
      weight: 0.3
      api_base: "https://api.anthropic.com/v1"
      api_key: ${ANTHROPIC_API_KEY}
    
    - name: "llama3"
      weight: 0.2
      api_base: "http://localhost:11434/v1"
      temperature: 0.8  # Per-model override
  
  temperature: 0.7
  max_tokens: 32000
```

### Reasoning Models (o1-series)

```yaml theme={null}
llm:
  models:
    - name: "o1"
      weight: 1.0
  reasoning_effort: "high"  # low, medium, or high
  max_tokens: 100000
```

## CLI Overrides

Override model settings from the command line:

```bash theme={null}
# Override model
skydiscover-run program.py evaluator.py -m gpt-5-mini

# Multiple models
skydiscover-run program.py evaluator.py -m "gpt-5,claude-opus-3"

# Override API base
skydiscover-run program.py evaluator.py --api-base http://localhost:8000/v1

# Both
skydiscover-run program.py evaluator.py \
  -m gemini-3-pro \
  --api-base https://generativelanguage.googleapis.com/v1beta/openai/
```

See `skydiscover/config.py:819-910` for override implementation.

## Environment Variables

Provider-specific API keys are automatically resolved:

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

# Anthropic
export ANTHROPIC_API_KEY="sk-ant-..."

# Google Gemini
export GEMINI_API_KEY="..."
export GOOGLE_API_KEY="..."  # Alternative

# DeepSeek
export DEEPSEEK_API_KEY="..."

# Mistral
export MISTRAL_API_KEY="..."

# Custom API base
export OPENAI_API_BASE="http://localhost:8000/v1"
export OPENAI_BASE_URL="http://localhost:8000/v1"  # Alternative
```

## Next Steps

<CardGroup cols={2}>
  <Card title="Prompt Configuration" icon="message" href="/config/prompt">
    Configure system messages and prompt templates
  </Card>

  <Card title="Search Configuration" icon="magnifying-glass" href="/config/search">
    Configure evolutionary search algorithms
  </Card>
</CardGroup>