Skip to content

Benchmark

JMLE2026-Bench (120th JMLE): correct answers, all questions

Science benchmarks measure how well models answer science questions. Higher scores are better.

Models measured
15
Scores recorded
27
Sources
1
IQ weight
2.01

Score distribution

326.1
Mean
353
Median
80.9
Std dev
191
10th percentile
386
90th percentile
112
Min
112393
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.01 for chat models, measured over 15 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Claude Opus 4.6· JMLE2026-Bench and IgakuQA119 data docs393
2Gemini 3.1 Pro Preview· JMLE2026-Bench and IgakuQA119 data docs393
3Claude Sonnet 4.6· JMLE2026-Bench and IgakuQA119 data docs391
4GPT 5.2· JMLE2026-Bench and IgakuQA119 data docs386
5Qwen3.5 397B A17B· JMLE2026-Bench and IgakuQA119 data docs382
6Qwen3.5 122B-A10B· JMLE2026-Bench and IgakuQA119 data docs381
7Qwen3.5 27B· JMLE2026-Bench and IgakuQA119 data docs381
8Qwen3.5 35B A3B· JMLE2026-Bench and IgakuQA119 data docs380
9GPT OSS 120Bhigh· JMLE2026-Bench and IgakuQA119 data docs374
10Qwen3.5 122B-A10Bno-think· JMLE2026-Bench and IgakuQA119 data docs366
11Qwen3.5 27Bno-think· JMLE2026-Bench and IgakuQA119 data docs365
12Qwen3.5 35B A3Bno-think· JMLE2026-Bench and IgakuQA119 data docs365
13GPT OSS 120Bmedium· JMLE2026-Bench and IgakuQA119 data docs363
14Qwen3 32B· JMLE2026-Bench and IgakuQA119 data docs353
15Qwen3.5 9B· JMLE2026-Bench and IgakuQA119 data docs346
16GPT OSS 120Blow· JMLE2026-Bench and IgakuQA119 data docs344
17GPT OSS 20bmedium· JMLE2026-Bench and IgakuQA119 data docs343
18GPT OSS 20bhigh· JMLE2026-Bench and IgakuQA119 data docs337
19Qwen3.5 9Bno-think· JMLE2026-Bench and IgakuQA119 data docs337
20Qwen3 32Bno-think· JMLE2026-Bench and IgakuQA119 data docs327
21Qwen3.5 4B· JMLE2026-Bench and IgakuQA119 data docs323
22GPT OSS 20blow· JMLE2026-Bench and IgakuQA119 data docs306
23Qwen3.5 4Bno-think· JMLE2026-Bench and IgakuQA119 data docs292
24Qwen3.5 2Bno-think· JMLE2026-Bench and IgakuQA119 data docs191
25Qwen3.5 2B· JMLE2026-Bench and IgakuQA119 data docs147
26Qwen3.5 0.8Bno-think· JMLE2026-Bench and IgakuQA119 data docs126
27Qwen3.5 0.8B· JMLE2026-Bench and IgakuQA119 data docs112

Data credits: JMLE2026-Bench and IgakuQA119 by Naoto Iwase; results licence not stated (code MIT, question data CC BY 4.0).