gpt-oss-20b
OpenAI · 2025-08-05
gpt-oss-20b is OpenAI's smaller open-weight reasoning model in the Apache 2.0-licensed gpt-oss family, sized for lower-latency, local, or specialized use cases. It is a text-only Mixture-of-Experts model with 21B total parameters and 3.6B active parameters per token, configurable reasoning effort, and tool-use support.