Coding benchmark # GSO leaderboard > GSO results for 31 AI models, led by Claude Fable 5.1 at 88.2%. What the benchmark measures, who runs it, and a source for every score. - Canonical page: https://noometry.com/benchmarks/gso-bench - Last updated: 2026-10-10 - Title: GSO Leaderboard (October 2026): Scores by Model | Noometry As of October 2026, Claude Fable 5.1 has the highest published GSO score on Noometry at 88.2%, out of 31 models with results. Last verified October 10, 2026 ## About GSO Software optimization tasks: the model must speed up real code bases while keeping them correct. - **Category:** [Coding](https://noometry.com/best/coding) - **Introduced:** 2025 - **Format:** Performance patch - **Unit:** Percent (random guessing ≈ 0%) - **Official site:** [gso-bench.github.io](https://gso-bench.github.io) ## Top 15 models Top models on GSO 1. Claude Fable 5.1 88.2% 2. GPT-6 Astra 79.4% 3. Claude Fable 5 78.4% 4. GPT-5.6 Sol 76.5% 5. Claude Opus 4.8 47.1% 6. Claude Opus 4.7 44.1% 7. Claude Opus 4.6 41.2% 8. GPT-5.5 40.2% 9. Claude Sonnet 5 37.3% 10. GPT-5.4 31.4% 11. GPT-5.2 27.4% 12. Claude Opus 4.5 26.5% 13. Gemini 3.1 Pro Preview 22.6% 14. Gemini 3 Pro 18.6% 15. Claude Sonnet 4.5 14.7% 16. 050100 Sponsored placements are available on pages like this one. [Advertise on Noometry](https://noometry.com/advertise) ## All results GSO results by model | # | Model | Provider | Score | Setting | Source | Date | | --- | --- | --- | --- | --- | --- | --- | | 1 | [Claude Fable 5.1](https://noometry.com/models/claude-fable-5-1) | [Anthropic](https://noometry.com/providers/anthropic) | 88.2% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 2 | [GPT-6 Astra](https://noometry.com/models/gpt-6-astra) | [OpenAI](https://noometry.com/providers/openai) | 79.4% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 3 | [Claude Fable 5](https://noometry.com/models/claude-fable-5) | [Anthropic](https://noometry.com/providers/anthropic) | 78.4% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 4 | [GPT-5.6 Sol](https://noometry.com/models/gpt-5-6-sol) | [OpenAI](https://noometry.com/providers/openai) | 76.5% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 5 | [Claude Opus 4.8](https://noometry.com/models/claude-opus-4-8) | [Anthropic](https://noometry.com/providers/anthropic) | 47.1% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 6 | [Claude Opus 4.7](https://noometry.com/models/claude-opus-4-7) | [Anthropic](https://noometry.com/providers/anthropic) | 44.1% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 7 | [Claude Opus 4.6](https://noometry.com/models/claude-opus-4-6) | [Anthropic](https://noometry.com/providers/anthropic) | 41.2% | high | [Epoch AI](https://epoch.ai/benchmarks) | | | 8 | [GPT-5.5](https://noometry.com/models/gpt-5-5) | [OpenAI](https://noometry.com/providers/openai) | 40.2% | xhigh | [Epoch AI](https://epoch.ai/benchmarks) | | | 9 | [Claude Sonnet 5](https://noometry.com/models/claude-sonnet-5) | [Anthropic](https://noometry.com/providers/anthropic) | 37.3% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 10 | [GPT-5.4](https://noometry.com/models/gpt-5-4) | [OpenAI](https://noometry.com/providers/openai) | 31.4% | xhigh | [Epoch AI](https://epoch.ai/benchmarks) | | | 11 | [GPT-5.2](https://noometry.com/models/gpt-5-2) | [OpenAI](https://noometry.com/providers/openai) | 27.4% | high | [Epoch AI](https://epoch.ai/benchmarks) | | | 12 | [Claude Opus 4.5](https://noometry.com/models/claude-opus-4-5) | [Anthropic](https://noometry.com/providers/anthropic) | 26.5% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 13 | [Gemini 3.1 Pro Preview](https://noometry.com/models/gemini-3-1-pro-preview) | [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 22.6% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 14 | [Gemini 3 Pro](https://noometry.com/models/gemini-3-pro) | [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 18.6% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 15 | [Claude Sonnet 4.5](https://noometry.com/models/claude-sonnet-4-5) | [Anthropic](https://noometry.com/providers/anthropic) | 14.7% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 16 | [GPT-5.1](https://noometry.com/models/gpt-5-1) | [OpenAI](https://noometry.com/providers/openai) | 13.7% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 17 | [Gemini 3 Flash Preview](https://noometry.com/models/gemini-3-flash-preview) | [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 9.8% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 18 | [o3](https://noometry.com/models/o3) | [OpenAI](https://noometry.com/providers/openai) | 8.8% | high | [Epoch AI](https://epoch.ai/benchmarks) | | | 19 | [Claude Opus 4](https://noometry.com/models/claude-opus-4) | [Anthropic](https://noometry.com/providers/anthropic) | 6.9% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 20 | [GPT-5](https://noometry.com/models/gpt-5) | [OpenAI](https://noometry.com/providers/openai) | 6.9% | high | [Epoch AI](https://epoch.ai/benchmarks) | | | 21 | [Claude Sonnet 4](https://noometry.com/models/claude-sonnet-4) | [Anthropic](https://noometry.com/providers/anthropic) | 4.9% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 22 | [Kimi K2 (Jul 2025)](https://noometry.com/models/kimi-k2) | [Moonshot AI](https://noometry.com/providers/moonshot) | 4.9% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 23 | [Qwen3-Coder 480B-A35B Instruct](https://noometry.com/models/qwen3-coder-480b-a35b-instruct) | [![](/logos/alibaba.svg) Alibaba (Qwen)](https://noometry.com/providers/alibaba) | 4.9% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 24 | [Claude 3.5 Sonnet](https://noometry.com/models/claude-3-5-sonnet) | [Anthropic](https://noometry.com/providers/anthropic) | 4.6% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 25 | [Gemini 2.5 Pro](https://noometry.com/models/gemini-2-5-pro) | [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 3.9% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 26 | [Gemini 2.5 Pro](https://noometry.com/models/gemini-2-5-pro) | [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 3.9% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 27 | [Claude 3.7 Sonnet](https://noometry.com/models/claude-3-7-sonnet) | [Anthropic](https://noometry.com/providers/anthropic) | 3.8% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 28 | [o4-mini](https://noometry.com/models/o4-mini) | [OpenAI](https://noometry.com/providers/openai) | 3.6% | high | [Epoch AI](https://epoch.ai/benchmarks) | | | 29 | [GLM-4.5-Air](https://noometry.com/models/glm-4-5-air) | [Z.ai (Zhipu)](https://noometry.com/providers/zai) | 2.9% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 30 | [o3-mini](https://noometry.com/models/o3-mini) | [OpenAI](https://noometry.com/providers/openai) | 1.3% | high | [Epoch AI](https://epoch.ai/benchmarks) | | | 31 | [GPT-4o](https://noometry.com/models/gpt-4o) | [OpenAI](https://noometry.com/providers/openai) | 0% | | [Epoch AI](https://epoch.ai/benchmarks) | | ## Compare the leaders - [Claude Fable 5.1 vs GPT-6 Astra](https://noometry.com/compare/claude-fable-5-1-vs-gpt-6-astra) - [Claude Fable 5.1 vs Claude Fable 5](https://noometry.com/compare/claude-fable-5-vs-claude-fable-5-1) - [Claude Fable 5.1 vs GPT-5.6 Sol](https://noometry.com/compare/claude-fable-5-1-vs-gpt-5-6-sol) - [Claude Fable 5.1 vs Claude Opus 4.8](https://noometry.com/compare/claude-fable-5-1-vs-claude-opus-4-8) - [GPT-6 Astra vs Claude Fable 5](https://noometry.com/compare/claude-fable-5-vs-gpt-6-astra) - [GPT-6 Astra vs GPT-5.6 Sol](https://noometry.com/compare/gpt-5-6-sol-vs-gpt-6-astra) ## Other coding benchmarks - [SWE-bench Verified](https://noometry.com/benchmarks/swe-bench-verified) - [DeepSWE](https://noometry.com/benchmarks/deepswe) - [FrontierCode](https://noometry.com/benchmarks/frontiercode) - [SWE-bench Verified (bash only)](https://noometry.com/benchmarks/swe-bench-bash-only) - [Aider Polyglot](https://noometry.com/benchmarks/aider-polyglot) - [LMArena WebDev](https://noometry.com/benchmarks/arena-webdev) - [CursorBench](https://noometry.com/benchmarks/cursorbench) - [SWE-bench Multilingual](https://noometry.com/benchmarks/swe-bench-multilingual) - [FrontierSWE](https://noometry.com/benchmarks/frontierswe) - [SciCode](https://noometry.com/benchmarks/scicode) - [WeirdML](https://noometry.com/benchmarks/weirdml) - [LMArena Coding](https://noometry.com/benchmarks/arena-coding) ## Frequently asked questions ### What does GSO measure? Software optimization tasks: the model must speed up real code bases while keeping them correct. ### Which model has the highest GSO score? As of October 2026, Claude Fable 5.1 has the highest published GSO score on Noometry at 88.2%, out of 31 models with results. ### What is the best open-weight model on GSO? Kimi K2 (Jul 2025) has the highest GSO accuracy among open-weight models at 4.9%, ranking 22 of 31 overall. ### Cite this page Noometry. (2026). GSO leaderboard. Retrieved October 10, 2026, from https://noometry.com/benchmarks/gso-bench Quote Noometry with a link back to this page. It is also available in [Markdown](https://noometry.com/md/benchmarks/gso-bench.md).