Agentic & Tool Use benchmark # Cybench leaderboard > Cybench results for 21 AI models, led by Claude Opus 4.6 at 93%. What the benchmark measures, who runs it, and a source for every score. - Canonical page: https://noometry.com/benchmarks/cybench - Last updated: 2026-10-10 - Title: Cybench Leaderboard (October 2026): Scores by Model As of October 2026, Claude Opus 4.6 has the highest published Cybench score on Noometry at 93%, out of 21 models with results. Last verified October 10, 2026 ## About Cybench Professional capture-the-flag cybersecurity tasks, solved without hints. - **Category:** [Agentic & Tool Use](https://noometry.com/best/agentic) - **Introduced:** 2024 - **Size:** 40 tasks - **Format:** CTF - **Unit:** Percent (random guessing ≈ 0%) - **Official site:** [cybench.github.io](https://cybench.github.io) ## Top 15 models Top models on Cybench 1. Claude Opus 4.6 93% 2. Claude Opus 4.5 82% 3. Claude Sonnet 4.5 60% 4. Grok 4 43% 5. Claude Opus 4.1 42% 6. Grok 4.1 39% 7. Claude Opus 4 38% 8. Claude Sonnet 4 35% 9. Grok 4 Fast 30% 10. o3-mini 22.5% 11. Claude 3.7 Sonnet 20% 12. Claude 3.5 Sonnet 17.5% 13. GPT-4.5 17.5% 14. GPT-4o 12.5% 15. Claude 3 Opus 10% 16. 050100 Sponsored placements are available on pages like this one. [Advertise on Noometry](https://noometry.com/advertise) ## All results Cybench results by model | # | Model | Provider | Score | Setting | Source | Date | | --- | --- | --- | --- | --- | --- | --- | | 1 | [Claude Opus 4.6](https://noometry.com/models/claude-opus-4-6) | [Anthropic](https://noometry.com/providers/anthropic) | 93% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 2 | [Claude Opus 4.5](https://noometry.com/models/claude-opus-4-5) | [Anthropic](https://noometry.com/providers/anthropic) | 82% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 3 | [Claude Sonnet 4.5](https://noometry.com/models/claude-sonnet-4-5) | [Anthropic](https://noometry.com/providers/anthropic) | 60% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 4 | [Grok 4](https://noometry.com/models/grok-4) | [xAI](https://noometry.com/providers/xai) | 43% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 5 | [Claude Opus 4.1](https://noometry.com/models/claude-opus-4-1) | [Anthropic](https://noometry.com/providers/anthropic) | 42% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 6 | [Grok 4.1](https://noometry.com/models/grok-4-1) | [xAI](https://noometry.com/providers/xai) | 39% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 7 | [Claude Opus 4](https://noometry.com/models/claude-opus-4) | [Anthropic](https://noometry.com/providers/anthropic) | 38% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 8 | [Claude Sonnet 4](https://noometry.com/models/claude-sonnet-4) | [Anthropic](https://noometry.com/providers/anthropic) | 35% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 9 | [Grok 4 Fast](https://noometry.com/models/grok-4-fast) | [xAI](https://noometry.com/providers/xai) | 30% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 10 | [o3-mini](https://noometry.com/models/o3-mini) | [OpenAI](https://noometry.com/providers/openai) | 22.5% | medium | [Epoch AI](https://epoch.ai/benchmarks) | | | 11 | [Claude 3.7 Sonnet](https://noometry.com/models/claude-3-7-sonnet) | [Anthropic](https://noometry.com/providers/anthropic) | 20% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 12 | [Claude 3.5 Sonnet](https://noometry.com/models/claude-3-5-sonnet) | [Anthropic](https://noometry.com/providers/anthropic) | 17.5% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 13 | [GPT-4.5](https://noometry.com/models/gpt-4-5) | [OpenAI](https://noometry.com/providers/openai) | 17.5% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 14 | [GPT-4o](https://noometry.com/models/gpt-4o) | [OpenAI](https://noometry.com/providers/openai) | 12.5% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 15 | [Claude 3 Opus](https://noometry.com/models/claude-3-opus) | [Anthropic](https://noometry.com/providers/anthropic) | 10% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 16 | [o1](https://noometry.com/models/o1) | [OpenAI](https://noometry.com/providers/openai) | 10% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 17 | [o1-mini](https://noometry.com/models/o1-mini) | [OpenAI](https://noometry.com/providers/openai) | 10% | medium | [Epoch AI](https://epoch.ai/benchmarks) | | | 18 | [Gemini 1.5 Pro (May 2024)](https://noometry.com/models/gemini-1-5-pro) | [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 7.5% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 19 | [Llama 3.1-405B](https://noometry.com/models/llama-3-1-405b) | [![](/logos/meta.svg) Meta](https://noometry.com/providers/meta) | 7.5% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 20 | [Mixtral 8x22B](https://noometry.com/models/mixtral-8x22b) | [![](/logos/mistral.svg) Mistral AI](https://noometry.com/providers/mistral) | 7.5% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 21 | [Llama 3-70B](https://noometry.com/models/llama-3-70b) | [![](/logos/meta.svg) Meta](https://noometry.com/providers/meta) | 5% | | [Epoch AI](https://epoch.ai/benchmarks) | | ## Compare the leaders - [Claude Opus 4.6 vs Claude Opus 4.5](https://noometry.com/compare/claude-opus-4-5-vs-claude-opus-4-6) - [Claude Opus 4.6 vs Claude Sonnet 4.5](https://noometry.com/compare/claude-opus-4-6-vs-claude-sonnet-4-5) - [Claude Opus 4.6 vs Grok 4](https://noometry.com/compare/claude-opus-4-6-vs-grok-4) - [Claude Opus 4.6 vs Claude Opus 4.1](https://noometry.com/compare/claude-opus-4-1-vs-claude-opus-4-6) - [Claude Opus 4.5 vs Claude Sonnet 4.5](https://noometry.com/compare/claude-opus-4-5-vs-claude-sonnet-4-5) - [Claude Opus 4.5 vs Grok 4](https://noometry.com/compare/claude-opus-4-5-vs-grok-4) ## Other agentic & tool use benchmarks - [Terminal-Bench](https://noometry.com/benchmarks/terminal-bench) - [APEX-Agents](https://noometry.com/benchmarks/apex-agents) - [Berkeley Function Calling Leaderboard](https://noometry.com/benchmarks/bfcl) - [OSWorld 2.0](https://noometry.com/benchmarks/osworld-2) - [GDPval](https://noometry.com/benchmarks/gdpval) - [Remote Labor Index](https://noometry.com/benchmarks/remote-labor-index) - [TheAgentCompany](https://noometry.com/benchmarks/the-agent-company) - [τ²-bench Airline](https://noometry.com/benchmarks/tau2-airline) - [τ²-bench Banking](https://noometry.com/benchmarks/tau2-banking) - [τ²-bench Retail](https://noometry.com/benchmarks/tau2-retail) - [τ²-bench Telecom](https://noometry.com/benchmarks/tau2-telecom) - [DeepResearch Bench](https://noometry.com/benchmarks/deepresearch-bench) ## Frequently asked questions ### What does Cybench measure? Professional capture-the-flag cybersecurity tasks, solved without hints. ### Which model has the highest Cybench score? As of October 2026, Claude Opus 4.6 has the highest published Cybench score on Noometry at 93%, out of 21 models with results. ### What is the best open-weight model on Cybench? Llama 3.1-405B has the highest Cybench accuracy among open-weight models at 7.5%, ranking 19 of 21 overall. ### Cite this page Noometry. (2026). Cybench leaderboard. Retrieved October 10, 2026, from https://noometry.com/benchmarks/cybench Quote Noometry with a link back to this page. It is also available in [Markdown](https://noometry.com/md/benchmarks/cybench.md).