The LLMs tab lets you explore all available language models, compare their capabilities, and configure which ones your agents can use.
Available providers
Provider
Models
OpenRouter
Access to 100+ models from all providers
OpenAI
GPT-4, GPT-4 Turbo, GPT-3.5
Anthropic
Claude Opus, Sonnet, Haiku
DeepSeek
DeepSeek Chat, DeepSeek Coder
Google
Gemini Pro, Gemini Flash
Model cards
Each model card shows:
Model name and provider
Pricing — per million input/output tokens
Context window — maximum token capacity
Capabilities — vision, function calling, JSON mode
Speed — relative latency rating
Selecting models
Models are assigned to agents during agent creation or editing. The LLMs tab helps you compare options before deciding.
OpenRouter is the default provider and gives you access to most models through a single API key. Configure your API key in Settings → Credentials.
Cost considerations
Track actual model costs in Analytics → LLM & Costs. The marketplace shows list pricing — your actual costs depend on usage volume and provider agreements.
