GSO

Measures: Software performance optimisation: in real repositories an agent must make a program faster; OPT@1 is the share of tasks where one attempt reaches at least 95% of the speedup a human expert achieved.

Maintainer: GSO team
Introduced: 2025
Status: Active
Saturation: Headroom
Typical provenance: Independent
In scoring set: Yes

Paper Official leaderboard

Leaderboard

Show independent results only

Every result we hold, whoever ran it. Only independent and third-party rows count towards a score.

Leaderboard
Model	OPT@1	Provenance	Source	Evaluated
Claude Opus 4.8 Anthropic · Unknown effort	47.1%	Third party	https://gso-bench.github.io/index.html via Epoch AI	—
Claude Opus 4.7 Anthropic · High effort	44.1%	Third party	https://gso-bench.github.io/index.html via Epoch AI	—
Claude Opus 4.7 Anthropic	44.1%	Third party	GSO Leaderboard via Epoch AI	—
Claude Opus 4.6 Anthropic · High effort	41.2%	Third party	GSO Leaderboard via Epoch AI	—
GPT-5.5 OpenAI · Xhigh effort	40.2%	Third party	https://gso-bench.github.io/index.html via Epoch AI	—
Claude Sonnet 5 Anthropic · Unknown effort	37.3%	Third party	https://gso-bench.github.io/index.html via Epoch AI	—
Claude Opus 4.6 Anthropic · Unknown effort	33.3%	Third party	GSO Leaderboard via Epoch AI	—
Claude Opus 4.6 Anthropic	33.3%	Third party	GSO Leaderboard via Epoch AI	—
GPT-5.4 OpenAI · Xhigh effort	31.4%	Third party	https://gso-bench.github.io/index.html via Epoch AI	—
GPT-5.2 OpenAI · High effort	27.4%	Third party	https://gso-bench.github.io/index.html via Epoch AI	—
Claude Opus 4.5 Anthropic	26.5%	Third party	GSO Leaderboard via Epoch AI	—
Claude Opus 4.5 Anthropic · Unknown effort	26.5%	Third party	GSO Leaderboard via Epoch AI	—
GPT-5.4 OpenAI · High effort	25.5%	Third party	https://gso-bench.github.io/index.html via Epoch AI	—
Gemini 3.1 Pro Google DeepMind	22.6%	Third party	https://gso-bench.github.io/index.html via Epoch AI	—
Gemini 3 Pro Google DeepMind	18.6%	Third party	https://gso-bench.github.io/index.html via Epoch AI	—
Claude Sonnet 4.5 Anthropic · Unknown effort	14.7%	Third party	GSO Leaderboard via Epoch AI	—
Claude Sonnet 4.5 Anthropic	14.7%	Third party	GSO Leaderboard via Epoch AI	—
GPT-5.1 OpenAI · Unknown effort	13.7%	Third party	https://gso-bench.github.io/index.html via Epoch AI	—
GPT-5.1 OpenAI · High effort	13.7%	Third party	GSO Leaderboard via Epoch AI	—
Gemini 3 Flash Google DeepMind	9.8%	Third party	https://gso-bench.github.io/index.html via Epoch AI	—
o3 OpenAI · High effort	8.8%	Third party	GSO Leaderboard via Epoch AI	—
Claude Opus 4 Anthropic	6.9%	Third party	GSO Leaderboard via Epoch AI	—
Claude Opus 4 Anthropic · Unknown effort	6.9%	Third party	GSO Leaderboard via Epoch AI	—
GPT-5 OpenAI · High effort	6.9%	Third party	https://gso-bench.github.io/index.html via Epoch AI	—
Claude Sonnet 4 Anthropic	4.9%	Third party	GSO Leaderboard via Epoch AI	—
Claude Sonnet 4 Anthropic · Unknown effort	4.9%	Third party	GSO Leaderboard via Epoch AI	—
Qwen3-Coder-480B-A35B-Instruct Alibaba Group	4.9%	Third party	https://gso-bench.github.io/index.html via Epoch AI	—
Claude 3.5 Sonnet Anthropic	4.6%	Third party	GSO Leaderboard via Epoch AI	—
Gemini 2.5 Pro Google DeepMind	3.9%	Third party	https://gso-bench.github.io/index.html via Epoch AI	—
Gemini 2.5 Pro Google DeepMind	3.9%	Third party	GSO Leaderboard via Epoch AI	—
Claude 3.7 Sonnet Anthropic	3.8%	Third party	GSO Leaderboard via Epoch AI	—
Claude 3.7 Sonnet Anthropic · Unknown effort	3.8%	Third party	GSO Leaderboard via Epoch AI	—
o4-mini OpenAI · High effort	3.6%	Third party	GSO Leaderboard via Epoch AI	—
GLM-4.5-Air Z.ai (Zhipu AI)	2.9%	Third party	GSO Leaderboard via Epoch AI	—
o3-mini OpenAI · Low effort	1.3%	Third party	GSO Leaderboard via Epoch AI	—
o3-mini OpenAI · High effort	1.3%	Third party	GSO Leaderboard via Epoch AI	—
GPT-4o OpenAI	0.0%	Third party	GSO Leaderboard via Epoch AI	—

Epoch AI, 'Capabilities & benchmarking', epoch.ai/benchmarks, CC BY 4.0.

Sources

Epoch AI