FeynmanBench - Topological Connectivity: leaderboard
FeynmanBench (2026) gives a multimodal model a machine-generated Feynman diagram (Standard Model QED, electroweak or QCD, tree level to two loops) with minimal labelling conventions and asks it to recover the full physical content of the diagram. Checkpoint 3: a vertex connection graph isomorphic to the diagram, judged by a three-model LLM panel; pass rate in percent, higher is better.
Metric: CP3 pass rate (%), a vertex connection graph isomorphic to the diagram, judged by a three-model LLM panel by majority vote, on FeynmanBench's Feynman-diagram tasks (a rendered diagram plus minimal conventions; Standard Model QED, electroweak and QCD diagrams from tree level to two loops), API models at temperature 0, mean of five runs; four models ran on subsets of 165 to 742 items, the others on about 2,000; higher is better. Source: arxiv.org. Saturation forecast: Around November 2027. 17 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 33.6 | #99 |
| 2 | Claude Sonnet 4.6 | 19.2 | #173 |
| 3 | Gemini 3 Flash | 16.8 | #175 |
| 4 | Qwen 3.5 Plus | 15.9 | #242 |
| 5 | Qwen 3.6 Plus | 15.2 | #238 |
| 6 | Seed 2.0 Pro | 13.3 | #222 |
| 7 | GLM-5V Turbo | 10.5 | #360 |
| 8 | Claude Opus 4.7 | 5.9 | #85 |
| 9 | Claude Opus 4.6 | 5.7 | #113 |
| 10 | Claude Opus 4.5 | 3.5 | #160 |
| 11 | GPT-5.2 | 3.3 | #185 |
| 12 | GPT-5.1 | 3.1 | #240 |
| 13 | GPT-5.4 | 3 | #135 |
| 14 | Kimi K2.5 | 1.4 | #266 |
| 15 | Claude Sonnet 4.5 | 1.1 | #261 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=feynmanbench-topological-connectivity · How It Works · Data refreshed daily, snapshot 2026-11-21.