ModelWatch
← Catalog

gpt-oss-120b

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.

open-weightslow-cost
Input price
$0.09 / 1M tokens
Output price
$0.36 / 1M tokens
Context window
131K tokens
Max output
118K tokens
Output speed
211 tokens/s
Time to first token
1.38s
Inputs
text
Open weights
Yes
OpenRouter id
openai/gpt-oss-120b
Listed
Aug 5, 2025

Benchmarks

Intelligence Index
3.8
Engineering Index
11.2

Speed and benchmarks for “gpt-oss-120b (High)” from Artificial Analysis. Prices, context window and listing date are live from OpenRouter.

Compare with others