FeynmanBench - Full Amplitude: 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 5: the complete scattering amplitude with Dirac structures, signs and symmetry factors, judged by a three-model LLM panel; pass rate in percent, higher is better.
Metric: CP5 pass rate (%), the complete scattering amplitude with Dirac structures, signs, fermion-loop and symmetry factors, judged by a three-model LLM panel, 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 2029. 17 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 7.8 | #99 |
| 2 | Claude Sonnet 4.6 | 3.3 | #173 |
| 3 | Qwen 3.6 Plus | 1.7 | #238 |
| 4 | Claude Opus 4.7 | 1.5 | #85 |
| 5 | Qwen 3.5 Plus | 1.5 | #242 |
| 6 | GLM-5V Turbo | 1.2 | #360 |
| 7 | Claude Opus 4.6 | 1 | #113 |
| 8 | GPT-5.4 | 0.9 | #135 |
| 9 | Seed 2.0 Pro | 0.9 | #222 |
| 10 | Qwen 3.6 Max | 0.2 | #193 |
| 11 | Claude Sonnet 4.5 | 0 | #261 |
| 12 | GPT-5.2 | 0 | #185 |
| 13 | Gemini 3 Flash | 0 | #175 |
| 14 | Kimi K2.5 | 0 | #266 |
| 15 | GPT-5.1 | 0 | #240 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=feynmanbench-full-amplitude · How It Works · Data refreshed daily, snapshot 2026-11-21.