# EEBench V1 core corpus (EEBench)

> An electrical-engineering agent benchmark maintained by atopile.

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

- Category: [Coding](/coding)
- Last updated: September 22, 2026 snapshot

## About EEBench

- Year: 2026
- Tasks: Electrical-engineering design tasks
- Format: Agent task score
- Difficulty: Professional electrical engineering
- Paper: [EEBench leaderboard](https://www.eebench.org/)

The public leaderboard exposes score, cost, and output-token fields. BenchLM stores the Grok 4.6 card comparison as display-only system evidence.

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

## Leaderboard (36 models)

| Rank | Model | Configuration | Creator | Score |
|------|-------|---------------|---------|-------|
| 1 | [Claude Opus 5.5](/models/claude-opus-5-5) | Claude Code | Anthropic | 75.0% |
| 2 | [GPT-6 Astra](/models/gpt-6-astra) | Codex CLI | OpenAI | 69.3% |
| 3 | [Claude Opus 5.5](/models/claude-opus-5-5) | Claude Code | Anthropic | 68.9% |
| 4 | [Grok 4.7](/models/grok-4-7) | Grok CLI | xAI | 64.0% |
| 5 | [Claude Opus 5](/models/claude-opus-5) | Claude Code | Anthropic | 61.6% |
| 6 | [Grok 4.6](/models/grok-4-6) | Grok CLI | xAI | 57.1% |
| 7 | [Claude Fable 5.1](/models/claude-fable-5-1) | Claude Code | Anthropic | 56.4% |
| 8 | [GPT-6 Sol](/models/gpt-6-sol) | Codex CLI | OpenAI | 56.3% |
| 9 | [Gemini 3.8 Flash](/models/gemini-3-8-flash) | Gemini CLI | Google | 55.4% |
| 10 | [Claude Fable 5](/models/claude-fable) | Claude Code | Anthropic | 54.3% |
| 11 | [Claude Opus 4.8](/models/claude-opus-4-8) | Claude Code | Anthropic | 51.4% |
| 12 | [Grok 4.5](/models/grok-4-5) | Grok CLI | xAI | 50.9% |
| 13 | [Gemini 3.1 Pro](/models/gemini-3-1-pro) | Gemini CLI | Google | 46.3% |
| 14 | [GPT-5.5](/models/gpt-5-5) | API tools | OpenAI | 42.3% |
| 15 | [Claude Sonnet 5](/models/claude-sonnet-5) | Claude Code | Anthropic | 40.3% |
| 16 | [GPT-5.6 Sol](/models/gpt-5-6-sol) | API tools | OpenAI | 39.4% |
| 17 | [GPT-5.4](/models/gpt-5-4) | API tools | OpenAI | 38.5% |
| 18 | [Kimi K3](/models/kimi-k3) | OpenRouter | Moonshot AI | 38.3% |
| 19 | [Gemini 3.6 Flash](/models/gemini-3-6-flash) | Gemini CLI | Google | 37.4% |
| 20 | [Claude Opus 4.8](/models/claude-opus-4-8) | Claude Code | Anthropic | 35.2% |
| 21 | [Gemini 3.5 Flash](/models/gemini-3-5-flash) | Gemini CLI | Google | 34.3% |
| 22 | [GPT-5.6 Sol Pro](https://www.eebench.org/) | OpenRouter | OpenAI | 32.2% |
| 23 | [GPT-6 Luna](/models/gpt-6-luna) | Codex CLI | OpenAI | 31.1% |
| 24 | [GPT-5.6 Terra](/models/gpt-5-6-terra) | API tools | OpenAI | 30.8% |
| 25 | [Claude Sonnet 4.6](/models/claude-sonnet-4-6) | Claude Code | Anthropic | 20.4% |
| 26 | [Gemini 3 Flash](/models/gemini-3-flash) | Gemini CLI | Google | 19.5% |
| 27 | [GPT-5.4 mini](/models/gpt-5-4-mini) | API tools | OpenAI | 18.3% |
| 28 | [Grok Build 0.1](/models/grok-build-0-1) | Grok CLI | xAI | 18.0% |
| 29 | [GLM-5.2](/models/glm-5-2) | OpenRouter | Z.AI | 16.7% |
| 30 | [Grok 4.3](/models/grok-4-3) | OpenRouter | xAI | 14.3% |
| 31 | [Kimi K2.7 Code](/models/kimi-k2-7-code) | OpenRouter | Moonshot AI | 12.8% |
| 32 | [Qwen3.6 Plus](/models/qwen3-6-plus) | OpenRouter | Alibaba | 12.7% |
| 33 | [DeepSeek V4 Pro 0813](/models/deepseek-v4-pro-0813) | OpenRouter | DeepSeek | 10.5% |
| 34 | [MiniMax M3](/models/minimax-m3) | OpenRouter | MiniMax | 9.2% |
| 35 | [Mistral Medium 3.5](https://www.eebench.org/) | OpenRouter | Mistral | 5.9% |
| 36 | [Claude Haiku 4.5](/models/claude-haiku-4-5) | Claude Code | Anthropic | 3.5% |

## FAQ

### What does EEBench measure?

An electrical-engineering agent benchmark maintained by atopile.

### Which model leads the published EEBench snapshot?

Claude Opus 5.5 currently leads the published EEBench snapshot with a score of 75.0%.

### How many models are evaluated on EEBench?

The September 22, 2026 snapshot contains 36 AI models.

### Does EEBench affect BenchLM's overall score?

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