Cerebras · gpt-oss

Cloud PlatformOpenAI compatible

Ultra-fast inference powered by wafer-scale compute chips, offering the fastest LLM speeds.

Active models
2+ 16 deprecated
Families
6
Headquarters🇺🇸 Sunnyvale, CA
Founded2016
Latest release2026-08-17
API Base URL
Modelsinference-docs.cerebras.ai/introduction
Capabilities
Model
OpenAI GPT OSSOSSgpt-oss-120b