Skip to content

Benchmark

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

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

Models measured
15
Scores recorded
27
Sources
1
IQ weight
2.03

Score distribution

250.9
Mean
273
Median
59.9
Std dev
148
10th percentile
296
90th percentile
89
Min
89300
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.03 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 docs300
2Claude Sonnet 4.6· JMLE2026-Bench and IgakuQA119 data docs298
3Gemini 3.1 Pro Preview· JMLE2026-Bench and IgakuQA119 data docs298
4GPT 5.2· JMLE2026-Bench and IgakuQA119 data docs296
5Qwen3.5 397B A17B· JMLE2026-Bench and IgakuQA119 data docs292
6Qwen3.5 27B· JMLE2026-Bench and IgakuQA119 data docs291
7Qwen3.5 35B A3B· JMLE2026-Bench and IgakuQA119 data docs290
8Qwen3.5 122B-A10B· JMLE2026-Bench and IgakuQA119 data docs289
9GPT OSS 120Bhigh· JMLE2026-Bench and IgakuQA119 data docs286
10Qwen3.5 122B-A10Bno-think· JMLE2026-Bench and IgakuQA119 data docs281
11Qwen3.5 27Bno-think· JMLE2026-Bench and IgakuQA119 data docs281
12Qwen3.5 35B A3Bno-think· JMLE2026-Bench and IgakuQA119 data docs280
13GPT OSS 120Bmedium· JMLE2026-Bench and IgakuQA119 data docs275
14Qwen3 32B· JMLE2026-Bench and IgakuQA119 data docs273
15Qwen3.5 9B· JMLE2026-Bench and IgakuQA119 data docs270
16GPT OSS 120Blow· JMLE2026-Bench and IgakuQA119 data docs263
17GPT OSS 20bmedium· JMLE2026-Bench and IgakuQA119 data docs262
18Qwen3.5 9Bno-think· JMLE2026-Bench and IgakuQA119 data docs261
19GPT OSS 20bhigh· JMLE2026-Bench and IgakuQA119 data docs258
20Qwen3.5 4B· JMLE2026-Bench and IgakuQA119 data docs253
21Qwen3 32Bno-think· JMLE2026-Bench and IgakuQA119 data docs251
22GPT OSS 20blow· JMLE2026-Bench and IgakuQA119 data docs235
23Qwen3.5 4Bno-think· JMLE2026-Bench and IgakuQA119 data docs228
24Qwen3.5 2Bno-think· JMLE2026-Bench and IgakuQA119 data docs148
25Qwen3.5 2B· JMLE2026-Bench and IgakuQA119 data docs121
26Qwen3.5 0.8Bno-think· JMLE2026-Bench and IgakuQA119 data docs104
27Qwen3.5 0.8B· JMLE2026-Bench and IgakuQA119 data docs89

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