OpenAI · Released Aug 5, 2025
gpt-oss-120b (High)
Openai's reasoning-optimized model featuring verified performance on the OpenRating benchmark index, balanced for high throughput and cost efficiency.
OpenRating Index
4.0
0–100 composite
Reasoning Index
5.4
0–100 composite
Cost / Task
$0.35
benchmark run
Output speed
204 tok/s
blended tok/s
Time to first token
10.2s
median latency
Input price
$0.15
per 1M tokens
Output price
$0.6
per 1M tokens
Context window
131K
tokens
Strengths
Intelligence Index score of 4.0 on independent evaluations
156.0 tokens/sec median generation throughput with sub-second response times
131k token context window supporting complex multi-document reasoning
How we rate gpt-oss-120b (High)
OpenRating scores every model on the same axes. The OpenRating Index is a weighted composite of reasoning, knowledge, and agentic benchmarks, normalised to a 0–100 scale. Speed is blended output measured in tokens per second under standardised load, with time-to-first-token reported separately. Prices are list input and output prices per million tokens, and Task Cost measures end-to-end workload spending.
OpenRating Index
4.0
Reasoning Index
5.4
Token Pricing
Input per 1M tokens
$0.15
Output per 1M tokens
$0.6
Cost of 1M input + 1M output
$0.75
Task Cost Economics
Empirical cost breakdown for a standard OpenRating Index benchmark task ($0.35 total).
Related models
Compare gpt-oss-120b (High) with the models it is most often measured against.