Many models, one platform
OpenAI, Anthropic, Google, xAI, Kimi, GLM. One agent, one memory. Your keys or ours.
OpenAI
4 modelsGPT-6.2 Sol
thinking1.1M context
OpenAI GPT-6.2 Sol for complex coding and agentic work, with a 1.05M context window.
GPT-6.2 Astra
thinking1.1M context
OpenAI frontier GPT-6.2 Astra. Top-tier intelligence for coding, computer use and professional work, with a 1.05M context window.
GPT-6.1 Terra
thinking1.1M context
Balanced GPT-6.1 model with strong reasoning at a fraction of the cost of Sol.
GPT-6 Luna
thinking1.1M context
Fast, cost-effective GPT-6 model for quick tasks and high-volume use cases.
Anthropic
4 modelsFable 5.2
thinking1M context
Anthropic's most capable model: state-of-the-art coding, vision, and computer use with a 1M context window and always-on adaptive thinking.
Opus 5.5
thinking1M context
Anthropic's Opus for long-running agentic coding and knowledge work, with a 1M context window and adaptive thinking.
Sonnet 5.5
thinking1M context
Latest Claude Sonnet with adaptive thinking, 1M context, and best speed-to-intelligence ratio.
Haiku 5.5
thinking1M context
Fastest and most compact Claude model, perfect for quick responses.
Gemini 4 Carbon
thinking1M context
Google's strongest Gemini 4 model for deep reasoning, with a 1M context window, vision, and PDF.
Gemini 4 Hydrogen
thinking1M context
Google's fast Gemini 4 model for agentic work and everyday coding, with a 1M context window, vision, PDF, and thinking.
xAI
1 modelGrok 4.7
thinking500K context
xAI flagship model for coding and agentic work, with a large context window.
Fireworks
5 modelsGLM-5.3 Flash
thinking1.0M context
Z.ai's GLM-5.3 Flash: a low-cost 1M-context model with vision, for routine coding work.
GLM-5.3
thinking1.0M context
Open-weight coding flagship (Z.ai) with a 1M context window, tuned for long-horizon agentic coding.
DeepSeek V4.1 Flash
thinking1.0M context
DeepSeek's V4.1 Flash on Fireworks: fast, low-cost reasoning and coding with vision and a 1M context window.
Kimi K3.1 Pro
thinking1.0M context
Moonshot's Kimi K3.1 Pro on the Fireworks lane: 1M context and native vision.
Qwen3.8 Max
thinking262K context
Alibaba's Qwen3.8 Max on Fireworks: built for long-horizon coding and autonomous repo work.