EEBench V1 core corpus (EEBench)

We show this table for reference; we do not rank on it.

An electrical-engineering agent benchmark maintained by atopile.

EEBench score on EEBench — September 22, 2026 snapshot

We mirror the published eebench score view for EEBench. Claude Opus 5.5 leads the public snapshot at 75.0%, followed by GPT-6 Astra (69.3%) and Claude Opus 5.5 (68.9%). We do not use these results to rank models overall.

1Proprietary

Claude Opus 5.5

Anthropic

Claude Code

75.0%

Overall 86.98Context 1M

2Proprietary

GPT-6 Astra

OpenAI

Codex CLI

69.3%

Overall 88.69Context 1.05M

3Proprietary

Claude Opus 5.5

Anthropic

Claude Code

68.9%

Overall 86.98Context 1M

36 modelsCodingCurrentDisplay onlyUpdated September 22, 2026 snapshot

EEBench score table (36 models)

Score

Claude Opus 5.5Anthropic · ClosedClaude Code

75.0%

GPT-6 AstraOpenAI · ClosedCodex CLI

69.3%

Claude Opus 5.5Anthropic · ClosedClaude Code

68.9%

Grok 4.7xAI · ClosedGrok CLI

64.0%

Claude Opus 5Anthropic · ClosedClaude Code

61.6%

Grok 4.6xAI · ClosedGrok CLI

57.1%

Claude Fable 5.1Anthropic · ClosedClaude Code

56.4%

GPT-6 SolOpenAI · ClosedCodex CLI

56.3%

Gemini 3.8 FlashGoogle · ClosedGemini CLI

55.4%

Claude Fable 5Anthropic · ClosedClaude Code

54.3%

Claude Opus 4.8Anthropic · ClosedClaude Code

51.4%

Grok 4.5xAI · ClosedGrok CLI

50.9%

Gemini 3.1 ProGoogle · ClosedGemini CLI

46.3%

GPT-5.5OpenAI · ClosedAPI tools

42.3%

Claude Sonnet 5Anthropic · ClosedClaude Code

40.3%

GPT-5.6 SolOpenAI · ClosedAPI tools

39.4%

GPT-5.4OpenAI · ClosedAPI tools

38.5%

Kimi K3Moonshot AI · ClosedOpenRouter

38.3%

Gemini 3.6 FlashGoogle · ClosedGemini CLI

37.4%

Claude Opus 4.8Anthropic · ClosedClaude Code

35.2%

Gemini 3.5 FlashGoogle · ClosedGemini CLI

34.3%

GPT-5.6 Sol ProOpenAIOpenRouter

32.2%

GPT-6 LunaOpenAI · ClosedCodex CLI

31.1%

GPT-5.6 TerraOpenAI · ClosedAPI tools

30.8%

Claude Sonnet 4.6Anthropic · ClosedClaude Code

20.4%

Gemini 3 FlashGoogle · ClosedGemini CLI

19.5%

GPT-5.4 miniOpenAI · ClosedAPI tools

18.3%

Grok Build 0.1xAI · ClosedGrok CLI

18.0%

GLM-5.2Z.AI · Open weightOpenRouter

16.7%

Grok 4.3xAI · ClosedOpenRouter

14.3%

Kimi K2.7 CodeMoonshot AI · Open weightOpenRouter

12.8%

Qwen3.6 PlusAlibaba · ClosedOpenRouter

12.7%

DeepSeek V4 Pro 0813DeepSeek · Open weightOpenRouter

10.5%

MiniMax M3MiniMax · Open weightOpenRouter

9.2%

Mistral Medium 3.5MistralOpenRouter

5.9%

Claude Haiku 4.5Anthropic · ClosedClaude Code

3.5%

How we show EEBench

We mirror the official EEBench V1 leaderboard generated on September 22, 2026 snapshot. The snapshot preserves all 36 published model-and-scaffold configurations together with uncertainty, attempt, cost, token, step, and duration fields.

EEBench is display only. The public results combine a model with scaffolds such as Claude Code, Grok CLI, Gemini CLI, API tools, or OpenRouter, so we keep them outside weighted model-only rankings.

Snapshot

36 configurations13 questions34 mapped rowsDisplay only

EEBench leaderboard Published leaderboard data

The published EEBench snapshot places Claude Opus 5.5 first at 75.0%. The third row is 6.1 points behind. The broader top-10 range is 20.7 points, so the table still separates the published systems.

36 models have been evaluated on EEBench. The benchmark falls in the Coding category. EEBench is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About EEBench

Year

2026

Tasks

Electrical-engineering design tasks

Format

Agent task score

Difficulty

Professional electrical engineering

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

EEBench leaderboard Public benchmark source

Freshness and provenance

Version

EEBench 2026

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

Questions

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 75.0% eebench score. BenchLM shows this benchmark for display only and does not use it in overall rankings.

How many models are evaluated on EEBench?

The September 22, 2026 snapshot snapshot contains 36 AI models.

Last updated: September 22, 2026 snapshot · mirrored from the public benchmark leaderboard

Know when it’s worth switching models

The model to choose, the cheaper alternative, and the release we would wait on.

Read a sample issue

Join 2,000+ readers.

One email each week. Unsubscribe anytime.