Overview
The smaller open reasoning tier, with 21B total and 3.6B active parameters per token, targeting local deployment and lower latency.
gpt-oss-20b
The smaller open reasoning tier, with 21B total and 3.6B active parameters per token, targeting local deployment and lower latency.
MODEL NUMBER296
- First public date
- 2025-08-05
- Organization
- OpenAI
- Family
- gpt-oss
01
Specifications
Only published specifications are shown.
02
Overview
03
Architecture & capabilities
Model capabilities
ReasoningCodingTool use
API features
Adjustable reasoning
04
Release history
gpt-oss-20b released model weights.
View source05
Related models
Select a card to view the related model.
06
Sources
Suggest a correction
Copy this template, add the field, proposed value, and primary source, and share it with the archive maintainer.