OpenAI

openai/…

Get an API key

Every OpenAI model we serve, with the numbers most providers leave out — the quantisation each endpoint actually runs at, measured first-token latency, and real uptime. Model ids are copyable: pass one straight to any OpenAI-compatible client.

On the roadmap

Not yet servable — these will not resolve through the API.

gpt-oss 120B

openai/gpt-oss-120b

Open-weight MoE with 5.1B active parameters. The best quality-per-dollar on the board — a single A100 serves it at MXFP4, which is where the undercut room is.

Other labs we serve

Qwen6DeepSeek3Ornith3Zz-ai3Infersia2Google1Mmeta-llama1Meta1Mminimax1Ttencent1Thinking Machines Lab1xAI1Xxiaomi1ZZeroEntropy1