OpenAI: gpt-oss-120b
OpenAIllm
Published August 5, 2025
About
gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases. It activates 5.1B parameters per forward pass and is optimized to run on a single H100 GPU with native MXFP4 quantization. The model supports configurable reasoning depth, full chain-of-thought access, and native tool use, including function calling, browsing, and structured output generation.
Try OpenAI: gpt-oss-120b
Test this model in the Sandbase Playground with your own prompts.
Open in Playground