Benchmark	Category	Top Score	Info	Self Reported	Source
Aider-Polyglot	code	0.34	High Reasoning Effort	Yes	Source
AIME 2024	math	0.96	High Reasoning Effort, With Tools	Yes	Source
AIME 2025	math	0.99	High Reasoning Effort, With Tools	Yes	Source
Codeforces	-	2516	High Reasoning Effort, With Tools	Yes	Source
EQ-Bench 3	-	800	-	No	Source
GPQA Diamond	general-knowledge	0.74	High Reasoning Effort, With Tools	Yes	Source
HealthBench	health	0.42	High Reasoning Effort	Yes	Source
HealthBench Concensus	health	0.85	Low Reasoning Effort	Yes	Source
HealthBench Hard	health	0.13	Medium Reasoning Effort	Yes	Source
Humanity's Last Exam	-	0.17	High Reasoning Effort, With Tools	Yes	Source
MMLU	-	0.85	High Reasoning Effort	Yes	Source
MMMLU	-	0.76	High Reasoning Effort, Average	Yes	Source
SWE-Bench	code	0.61	High Reasoning Effort	Yes	Source
Tau Bench (Airline)	-	0.43	Medium Reasoning Effort	Yes	Source
Tau Bench (Retail)	-	0.55	High Reasoning Effort	Yes	Source

Use the selector to switch benchmarks and see how this model stacks up against its closest competitors.

Compare this model with the leading peers for the selected benchmark.

Benchmark

0.76

Rank #5/5

5 models

Showing 5 models around the selected model (out of 5 total).

Preparing your page

GPT OSS 20b