Submitting stores these access-request details so we can evaluate and contact you. See the privacy policy.

Current direct API

  • OpenAI-compatible chat completions
  • Qwen3.8-27B with a tested 262,144-token context
  • Streaming, tool calls, structured output, and reasoning
  • US-hosted inference with no persistent prompt or completion storage
  • Account-scoped keys and per-request usage reporting
Pilot terms: access is reviewed manually, capacity is limited, and no uptime SLA is offered. Current rates are published on Models & pricing.
Already invited? Sign in to your dashboard to create an API key.