claude-haiku-5-5
Designed for high-volume summarization, classification and bounded subagent tasks, with adjustable effort.
Pricing & details →AI model API pricing and integration.
Compare catalog identifiers, protocol support and published pricing snapshots. Verify your account’s available channels before sending requests.
Compare prices and estimate a request
Published price snapshot, USD per million tokens. Account availability and the actual bill must be checked separately. Route names retain their configured labels.
Base-tier rates are available below without JavaScript. Enable JavaScript to filter or calculate.
Reviewed model capabilities
claude-opus-5-5
Targets complex coding and sustained tasks requiring judgment. Evaluate accepted outcomes and unintended changes.
Pricing & details →claude-sonnet-5-5
Targets everyday development, document work and well-scoped agent workflows. Compare quality and cost across complete tasks.
Pricing & details →gemini-4-argon
A stable vendor flagship for coding, scientific reasoning and multi-step tasks. Documented inputs include text, images, video, audio and PDF, with text output.
Pricing & details →gpt-6-astra
For complex coding, research and tasks with interacting constraints. The vendor supports text/image input and text output; tools require application setup.
Pricing & details →gpt-6.1-sol
Evaluate routine coding and document work against quality and cost targets. OpenAI positions it as a capability/cost balance; no independent ranking is claimed.
Pricing & details →kimi-k3
Moonshot’s flagship for long materials, visual understanding and demanding coding. Evaluate conversation integrity, output quality and cache costs.
Pricing & details →