Skip to content

Benchmark

Vals MedQA

Science benchmarks measure how well models answer science questions. Scores are percentages.

Models measured
83
Scores recorded
88
Sources
1
IQ weight
2.63

Score distribution

84.3%
Mean
90.2%
Median
15.2%
Std dev
62.6%
10th percentile
95.9%
90th percentile
2.7%
Min
2.7%96.5%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.63 for chat models, measured over 83 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1o1· Vals AI96.5%
2GPT 5.1· Vals AI96.4%
3Gemini 3.1 Pro Preview· Vals AI96.4%
4GPT 5· Vals AI96.3%
5GPT 5.4· Vals AI96.1%
6GPT-5 mini· Vals AI96.1%
7OpenAI o3· Vals AI96.1%
8Gemini 3 Pro Preview· Vals AI96%
9o4 Mini· Vals AI96%
10Claude Opus 4.5· Vals AI95.9%
11Gemini 3 Flash· Vals AI95.8%
12Claude Opus 4.6· Vals AI95.4%
13Qwen3.5 Plusthinking· Vals AI95.2%
14o3 Mini· Vals AI94.8%
15Claude Sonnet 4.5· Vals AI94.7%
16Grok 4.20· Vals AI94.6%
17Kimi K2.5thinking· Vals AI94.4%
18GLM 5thinking· Vals AI94.3%
19GPT 5.2· Vals AI94.1%
20DeepSeek V3.2· Vals AI93.9%
21GLM 4.7· Vals AI93.7%
22Claude Opus 4.1· Vals AI93.6%
23GPT-5 nano· Vals AI93.3%
24Gemini 2.5 Pro· Vals AI93.1%
25O1 Preview· Vals AI93%
26Claude Opus 4· Vals AI92.9%
27Claude Sonnet 4thinking· Vals AI92.7%
28Kimi K2 Thinking· Vals AI92.6%
29MiniMax M2.5· Vals AI92.5%
30Grok 4· Vals AI92.5%
31Grok 2· Vals AI92.3%
32GLM 4.6· Vals AI92.2%
33Grok 4.1 Fast· Vals AI92.1%
34Grok 4 Fast· Vals AI92.1%
35Claude Sonnet 4.6· Vals AI92.1%
36gemini-2.5-flash-preview-09-2025· Vals AI91.4%
37GPT OSS 120B· Vals AI91.4%
38GPT 4.1· Vals AI91.2%
39gemini-2.5-flash-preview-09-2025thinking· Vals AI91.2%
40MiniMax M2.1· Vals AI91.2%
41gemini-2.5-flash-preview-04-17thinking· Vals AI91%
42DeepSeek R1· Vals AI90.8%
43Qwen 3 235B A22B· Vals AI90.6%
44Claude Sonnet 4· Vals AI90.4%
45Claude Sonnet 3.7thinking· Vals AI90.2%
46O1 Mini· Vals AI90.2%
47Grok 3 Mini Fast Betahigh, reasoning· Vals AI90.1%
48GLM 4.5· Vals AI90%
49Magistral Medium· Vals AI89.5%
50Gemini 2.5 Flash Lite Preview 09-2025thinking· Vals AI88.9%

Showing 1–50 of 88 results.