Skip to main content
Agent commands. Your AI agent uses this node in workflows. You don’t configure it directly.
The LLM node calls AI models using Simon Willison’s llm library. It supports multiple providers (OpenAI, Anthropic, Google, local models) through a unified interface.

​
Parameters

ParameterTypeRequiredDefaultDescription
promptstrYes-Text prompt to send to the model
modelstrNogpt-5.2Model identifier
systemstrNo-System prompt for behavior guidance
temperaturefloatNo1.0Sampling temperature (0.0-2.0)
max_tokensintNo-Maximum response tokens
imageslistNo[]Image URLs or file paths for vision models

​
Output

KeyTypeDescription
responseanyModel’s response (auto-parsed JSON or string)
llm_usagedictToken usage metrics
errorstrError message (only present on failure)

​
Token usage structure

{
  "model": "gpt-5.2",
  "input_tokens": 150,
  "output_tokens": 89,
  "total_tokens": 239,
  "cache_creation_input_tokens": 0,
  "cache_read_input_tokens": 0
}

​
Model support

These providers are included with pflow - just set your API key:
ProviderExample models
OpenAIgpt-5.2, gpt-5.1, gpt-4o
Anthropicclaude-opus-4-5, claude-sonnet-4-5, claude-haiku-4-5
Googlegemini-3.0-pro, gemini-2.5-flash
# Set API keys (stored in ~/.pflow/settings.json)
pflow settings set-env OPENAI_API_KEY "sk-..."
pflow settings set-env ANTHROPIC_API_KEY "sk-ant-..."
pflow settings set-env GEMINI_API_KEY "..."
Run llm models to see all available models on your system.

​
Extending with plugins

pflow uses Simon Willison’s llm library, which supports plugins for additional providers and local models.

​
Installing plugins

If you installed pflow with uv tool, include plugins during installation: 
uv tool install --with llm-openrouter pflow-cli
If you installed with pipx, use inject to add plugins: 
pipx inject pflow-cli llm-openrouter
Plugins must be installed in pflow’s environment. Running llm install separately won’t work with isolated installations.
PluginInstall flagUse case
llm-openrouter--with llm-openrouterAccess Claude, GPT, Llama, Mistral via OpenRouter
llm-ollama--with llm-ollamaRun models locally with Ollama
After installing, set up credentials: 
# OpenRouter - get key from https://openrouter.ai/keys
llm keys set openrouter

# Ollama - no API key needed
brew install ollama
ollama serve
ollama pull llama3.2

​
Using plugin models

OpenRouter models use the openrouter/provider/model format: 
{
  "type": "llm",
  "params": {
    "model": "openrouter/anthropic/claude-sonnet-4-5",
    "prompt": "Summarize this"
  }
}
Ollama models use the model name directly: 
{
  "type": "llm",
  "params": {
    "model": "llama3.2",
    "prompt": "Summarize this"
  }
}
See the llm plugin directory for more providers including Mistral, Bedrock, and other local model options.

​
Automatic JSON parsing

The node automatically detects and parses JSON responses: 
{
  "id": "analyze",
  "type": "llm",
  "params": {
    "prompt": "List 3 colors as JSON array",
    "model": "gpt-4o-mini"
  }
}
If the response is valid JSON (including markdown-wrapped JSON), response will be the parsed object/array. Otherwise, it’s the raw string.

​
Image support

For vision-capable models, pass image URLs or local file paths: 
{
  "id": "describe",
  "type": "llm",
  "params": {
    "prompt": "What's in this image?",
    "model": "gpt-5.2",
    "images": ["photo.jpg"]
  }
}
Supported formats: JPEG, PNG, GIF, WebP, PDF Images can be:
  • Local file paths: photo.jpg, /path/to/image.png
  • URLs: https://example.com/image.jpg

​
Examples

​
Basic prompt

{
  "nodes": [
    {
      "id": "summarize",
      "type": "llm",
      "params": {
        "prompt": "Summarize: ${read.content}",
        "model": "gpt-4o-mini"
      }
    }
  ]
}

​
With system prompt

{
  "nodes": [
    {
      "id": "translate",
      "type": "llm",
      "params": {
        "system": "You are a translator. Respond only with the translation.",
        "prompt": "Translate to Spanish: ${input.text}",
        "temperature": 0.3
      }
    }
  ]
}

​
Structured output

{
  "nodes": [
    {
      "id": "extract",
      "type": "llm",
      "params": {
        "system": "Extract entities as JSON with keys: people, places, organizations",
        "prompt": "${document.content}"
      }
    }
  ]
}
The response will be automatically parsed if it’s valid JSON.

​
Image analysis

{
  "nodes": [
    {
      "id": "analyze",
      "type": "llm",
      "params": {
        "prompt": "Describe the main elements in this image",
        "model": "gpt-5.2",
        "images": ["${file_path}"]
      }
    }
  ]
}

​
Error handling

ErrorCauseSolution
Unknown modelModel ID not recognizedRun llm models to see available models
API key requiredMissing credentialsSet up with llm keys set <provider> or env var
Rate limitToo many requestsWait and retry automatically (built-in retry)
The node retries transient failures automatically (3 attempts, 1 second wait).