GPT-OSS 120B API

The GPT-OSS 120B API provides access to an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI, built for high-reasoning, agentic workflows, and general-purpose production use cases. Its efficient MoE design activates only 5.1B parameters per forward pass and is optimized to run on a single H100 GPU using native MXFP4 quantization for exceptional performance and cost efficiency.

Released under the Apache 2.0 license, the GPT-OSS 120B API delivers full model transparency with configurable chain-of-thought reasoning and native tool use, including function calling and browsing. This enables organizations to customize behavior while maintaining complete data control for secure, private, and compliant deployments.