# Vals SRE Bench (formerly ReverseEngBench) (SRE Bench)

> A contamination-resistant agent benchmark for reverse engineering real-world binaries.

Canonical page: https://benchlm.ai/benchmarks/valsreverseengbench

- Category: [Agentic](/agentic)
- Last updated: October 14, 2026

Evaluation results published by [Vals AI](https://www.vals.ai/benchmarks/srebench). Published leaderboard aggregates and evaluation settings only. Dataset contents are not included.

## About SRE Bench

- Year: 2026
- Tasks: Real-world binary reverse-engineering tasks
- Format: Fully solved rate and capability score
- Difficulty: Agentic reverse engineering
- Paper: [SRE Bench](https://www.vals.ai/benchmarks/srebench)

Vals published this board as ReverseEngBench until September 2026 and now lists it as SRE Bench under the same sre_bench benchmark id. The public table reports fully solved and partial-capability scores through an external agent runner. We keep it display only because Vals marks the dataset private and the result measures the complete agent setup.

SRE Bench is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Leaderboard (35 models)

| Rank | Model | Configuration | Creator | Score |
|------|-------|---------------|---------|-------|
| 1 | [GPT-6 Astra](/models/gpt-6-astra) | max reasoning | OpenAI | 56.87% |
| 2 | [GPT-6.1 Sol](/models/gpt-6-1-sol) | max reasoning | OpenAI | 50.76% |
| 3 | [Gemini 4 Argon](/models/gemini-4-argon) | high reasoning | Google | 44.27% |
| 4 | [Claude Opus 5.5](/models/claude-opus-5-5) | — | Anthropic | 33.59% |
| 5 | [GPT Daybreak Blue](https://www.vals.ai/models/openai_gpt-daybreak-blue) | max reasoning | OpenAI | 31.68% |
| 6 | [GPT-5.6 Sol](/models/gpt-5-6-sol) | max reasoning | OpenAI | 30.53% |
| 7 | [GPT-6 Sol](/models/gpt-6-sol) | max reasoning | OpenAI | 30.53% |
| 8 | [Claude Sonnet 5.5](/models/claude-sonnet-5-5) | — | Anthropic | 30.15% |
| 9 | [Claude Fable 5.1](/models/claude-fable-5-1) | — | Anthropic | 22.90% |
| 10 | [Claude Opus 5.5 Range](https://www.vals.ai/models/anthropic_claude-opus-5-5-range) | — | Anthropic | 21.76% |
| 11 | [Claude Opus 5](/models/claude-opus-5) | — | Anthropic | 12.21% |
| 12 | [Muse Wonder Full](/models/muse-wonder-full) | max reasoning | Meta | 11.83% |
| 13 | [Gemini 3.8 Flash](/models/gemini-3-8-flash) | high reasoning | Google | 11.07% |
| 14 | [GPT-5.6 Terra](/models/gpt-5-6-terra) | max reasoning | OpenAI | 9.54% |
| 15 | [GLM-5.3](/models/glm-5-3) | max reasoning | Z.AI | 9.16% |
| 16 | [Claude Haiku 5.5](/models/claude-haiku-5-5) | — | Anthropic | 5.72% |
| 17 | [Gemini 3.7 Flash](/models/gemini-3-7-flash) | high reasoning | Google | 4.58% |
| 18 | [MiMo-V2.6-Flash](/models/mimo-v2-6-flash) | — | Xiaomi | 4.20% |
| 19 | [Grok 4.7](/models/grok-4-7) | xhigh reasoning | xAI | 3.82% |
| 20 | [GPT-5.5](/models/gpt-5-5) | xhigh reasoning | OpenAI | 3.82% |
| 21 | [Muse Spark 1.3 Max](https://www.vals.ai/models/meta_muse_spark_1_3_max) | max reasoning | Meta | 3.82% |
| 22 | [MiMo-V2.6-Pro](/models/mimo-v2-6-pro) | — | Xiaomi | 3.05% |
| 23 | [GPT-6 Luna](/models/gpt-6-luna) | max reasoning | OpenAI | 2.67% |
| 24 | [Hy4 preview](/models/hy4-preview) | — | Tencent | 2.29% |
| 25 | [DeepSeek V4 Pro 0813](/models/deepseek-v4-pro-0813) | max reasoning | DeepSeek | 1.91% |
| 26 | [GPT-5.6 Luna](/models/gpt-5-6-luna) | max reasoning | OpenAI | 1.53% |
| 27 | [DeepSeek V4.1 Flash 1017](/models/deepseek-v4-1-flash) | max reasoning | DeepSeek | 0.76% |
| 28 | [GLM-5.3-Flash](/models/glm-5-3-flash) | max reasoning | Z.AI | 0.76% |
| 29 | [Grok 4.5](/models/grok-4-5) | high reasoning | xAI | 0.76% |
| 30 | [Claude Sonnet 5](/models/claude-sonnet-5) | — | Anthropic | 0.38% |
| 31 | [Step 5 Preview](/models/step-5-preview) | — | StepFun | 0.38% |
| 32 | [GLM-5.2](/models/glm-5-2) | max reasoning | Z.AI | 0.00% |
| 33 | [Muse Spark 1.2](/models/muse-spark-1-2) | xhigh reasoning | Meta | 0.00% |
| 34 | [Gemini 3.1 Pro Preview](https://www.vals.ai/models/google_gemini-3.1-pro-preview) | high reasoning | Google | 0.00% |
| 35 | [Inkling](/models/inkling) | 0.99 reasoning | Thinking Machines Lab | 0.00% |

## FAQ

### What does SRE Bench measure?

A contamination-resistant agent benchmark for reverse engineering real-world binaries.

### Which model leads the published SRE Bench snapshot?

GPT-6 Astra currently leads the published SRE Bench snapshot with a score of 56.87%.

### How many models are evaluated on SRE Bench?

The October 14, 2026 contains 35 AI models.

### Does SRE Bench affect BenchLM's overall score?

Not directly. SRE Bench is still displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.