Available models
27 curated models across 11 providers. One API. Zero lock-in.
OpenAI
GPT-4o
Multimodal flagship with 128K context, tool use, and structured outputs.
gpt-4oJump to detailsGPT-4o Mini
Cost-efficient small model for high-throughput, low-latency tasks.
gpt-4o-miniJump to detailso1
Reasoning model with chain-of-thought for math, coding, and science.
o1Jump to detailso3-mini
Fast, affordable reasoning model for STEM and structured problem-solving.
o3-miniJump to details
Anthropic
Claude 3.5 Sonnet
Balanced speed and intelligence with 200K context window.
claude-3-5-sonnetJump to detailsClaude 3.5 Haiku
Fastest Anthropic model for lightweight tasks and agents.
claude-3-5-haikuJump to detailsClaude 3 Opus
Maximum intelligence for complex analysis and long-form generation.
claude-3-opusJump to details
Gemini 2.0 Flash
Low-latency multimodal model with 1M token context.
gemini-2.0-flashJump to detailsGemini 2.0 Pro
Google's most capable model for complex reasoning and code.
gemini-2.0-proJump to detailsGemini 2.5 Pro
Experimental reasoning model with extended thinking capabilities.
gemini-2.5-proJump to detailsGemini Chrome
Google Gemini 2.5 Pro via Nimbus session bridge. 1M input tokens, 65K output tokens, 10-15s typical latency. $2/M input, $10/M output raw wholesale.
gemini-chromeJump to details