# CyberGym-E2E-AA leaderboard

Agentic benchmark

> Every figure below is reproduced as its upstream source published it: nothing is modelled, estimated, interpolated or converted. `Unavailable` means no source published the value — it is never a zero. Each value carries its evidence state and the date it was observed.

Page: https://modelchoice.dev/leaderboards/benchmarks/agentic-aa-cyber-gym-e2e

Artificial Analysis CyberGym-E2E-AA. Every model the catalog carries a published CyberGym-E2E-AA value for, ranked by that value.

## Benchmark record

| Field | Value |
| --- | --- |
| Category | Agentic |
| Measure | Pass@1 |
| Tasks | 131 end-to-end vulnerability reproduction and patching tasks |
| Difficulty | Real-world cybersecurity |
| Published by | [CyberGym-E2E-AA Benchmark Leaderboard](https://artificialanalysis.ai/evaluations/cybergym-e2e-aa) |

Artificial Analysis' implementation of Berkeley RDI's CyberGym-E2E: agents find a real memory-safety vulnerability in a C/C++ open-source project, write a proof-of-concept input that crashes it, and patch it without breaking the project's tests.

## CyberGym-E2E-AA ranking

21 models with a published CyberGym-E2E-AA value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.

| Rank | Model | Provider | Pass@1 | Evidence |
| --- | --- | --- | --- | --- |
| 1 | [Kimi K3.1 Pro](https://modelchoice.dev/models/kimi-3-1-pro) | Moonshot AI | 94.7 | Observed 2026-11-21 · source benchlm:models |
| 2 | [Mistral Large 4](https://modelchoice.dev/models/mistral-large-4) | Mistral | 79.4 | Observed 2026-11-21 · source benchlm:models |
| 3 | [MiMo-V2.6-Pro](https://modelchoice.dev/models/mimo-v2-6-pro) | Xiaomi | 78.6 | Observed 2026-11-21 · source benchlm:models |
| 4 | [GPT-6 Luna](https://modelchoice.dev/models/gpt-6-luna) | OpenAI | 77.9 | Observed 2026-11-21 · source benchlm:models |
| 5 | [Step 5 Preview](https://modelchoice.dev/models/step-5-preview) | StepFun | 74.8 | Observed 2026-11-21 · source benchlm:models |
| 6 | [GLM-5.3-Flash](https://modelchoice.dev/models/glm-5-3-flash) | Z.AI | 74.0 | Observed 2026-11-21 · source benchlm:models |
| 6 | [Grok 4.7](https://modelchoice.dev/models/grok-4-7) | xAI | 74.0 | Observed 2026-11-21 · source benchlm:models |
| 8 | [Kimi K3](https://modelchoice.dev/models/kimi-3) | Moonshot AI | 58.0 | Observed 2026-11-21 · source benchlm:models |
| 9 | [Beam 1.0 501B-A23B](https://modelchoice.dev/models/reflection-beam) | Reflection AI | 45.0 | Observed 2026-11-21 · source benchlm:models |
| 10 | [GLM-5.3](https://modelchoice.dev/models/glm-5-3) | Z.AI | 29.0 | Observed 2026-11-21 · source benchlm:models |
| 11 | [MiniMax M3](https://modelchoice.dev/models/minimax-m3) | MiniMax | 24.4 | Observed 2026-11-21 · source benchlm:models |
| 12 | [DeepSeek V4.1 Flash 1017](https://modelchoice.dev/models/deepseek-v4-1-flash) | DeepSeek | 22.9 | Observed 2026-11-21 · source benchlm:models |
| 13 | [Nemotron 3 Ultra](https://modelchoice.dev/models/nemotron-3-ultra-500b) | NVIDIA | 10.7 | Observed 2026-11-21 · source benchlm:models |
| 14 | [Gemini 3.8 Flash](https://modelchoice.dev/models/gemini-3-8-flash) | Google | 8.4 | Observed 2026-11-21 · source benchlm:models |
| 15 | [Claude Fable 5.1](https://modelchoice.dev/models/claude-fable-5-1) | Anthropic | 0.8 | Observed 2026-11-21 · source benchlm:models |
| 15 | [Claude Opus 5.5](https://modelchoice.dev/models/claude-opus-5-5) | Anthropic | 0.8 | Observed 2026-11-21 · source benchlm:models |
| 17 | [GPT-6 Astra](https://modelchoice.dev/models/gpt-6-astra) | OpenAI | 0.0 | Observed 2026-11-21 · source benchlm:models |
| 17 | [GPT-6 Sol](https://modelchoice.dev/models/gpt-6-sol) | OpenAI | 0.0 | Observed 2026-11-21 · source benchlm:models |
| 17 | [GPT-6.2 Astra](https://modelchoice.dev/models/gpt-6-2-astra) | OpenAI | 0.0 | Observed 2026-11-21 · source benchlm:models |
| 17 | [GPT-6.2 Sol](https://modelchoice.dev/models/gpt-6-2-sol) | OpenAI | 0.0 | Observed 2026-11-21 · source benchlm:models |
| 17 | [Qwen3.8-27B](https://modelchoice.dev/models/qwen3-8-27b) | Alibaba | 0.0 | Observed 2026-11-21 · source benchlm:models |

Rows are ordered by the value the source published, highest first. The source does not state whether a higher value is the better result, so this document does not either: for a benchmark that measures a rate of failure, read the table from the bottom.
