QCalEval: leaderboard

Calibration plot understanding on QCalEval (2026), 243 quantum computing calibration plots from 22 experiment families of superconducting-qubit and neutral-atom experiments, answered zero-shot with the experiment background. Mean of the six question-type scores (description, outcome, significance, fit, parameters, diagnosis), from 0 to 100; higher is better.

Metric: Mean score (0-100) over the six question types (description, outcome, significance, fit, parameters, diagnosis), on the 243 quantum-calibration examples of QCalEval (22 experiment families, superconducting qubits and neutral atoms), zero-shot with the family background text, greedy decoding; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 18 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Ising-Calibration-1-35B-A3B74.7
2Gemini 3.1 Pro (Preview)72.3#99
3Gemma 4 31B (IT)68.8#350
4Gemini 3.1 Flash Lite68.2#371
5Claude Opus 4.667.8#113
6Claude Sonnet 4.666.5#173
7GPT-5.464.6#135
8Qwen 3.5 397B A17B58.6#264
9Qwen 3.5 27B58.5#375
10Qwen 3.5 122B A10B57.1#346
11GPT-5.4 Mini55.7#406
12Qwen 3.5 35B A3B55.5#492
13Qwen 3.5 9B53#828
14Claude Haiku 4.550.5#583
15InternVL3-78B48.2#759

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=qcaleval · How It Works · Data refreshed daily, snapshot 2026-11-21.