Skip to content

Benchmark

Forecastbench Dataset score

Scores are percentages.

Models measured
85
Scores recorded
95
Sources
1
IQ weight
1.74

Score distribution

58.5%
Mean
59.8%
Median
3.7%
Std dev
54.9%
10th percentile
61.7%
90th percentile
35%
Min
35%63.7%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.74 for chat models, measured over 85 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1GPT-6 Solhigh· Epoch AI Benchmarking Hub data63.7%
2GPT 5.5high· Epoch AI Benchmarking Hub data63.3%
3GPT-6 Astrahigh· Epoch AI Benchmarking Hub data63.1%
4Claude Opus 5.5high· Epoch AI Benchmarking Hub data62.4%
5Muse Spark 1.3max· Epoch AI Benchmarking Hub data62.3%
6Claude Opus 5max· Epoch AI Benchmarking Hub data62.1%
7Gemini 3.7 Flashhigh· Epoch AI Benchmarking Hub data62%
8Claude Fable 5.1medium· Epoch AI Benchmarking Hub data61.8%
9Gemini 3.6 Flashmedium· Epoch AI Benchmarking Hub data61.8%
10GPT-5.6 Solmedium· Epoch AI Benchmarking Hub data61.7%
11Claude Opus 4.8max· Epoch AI Benchmarking Hub data61.6%
12Grok 4.6xhigh· Epoch AI Benchmarking Hub data61.5%
13Claude Fable 5max· Epoch AI Benchmarking Hub data61.4%
14GPT-5.6 Solpro / max· Epoch AI Benchmarking Hub data61.4%
15Grok 4.5high· Epoch AI Benchmarking Hub data61.3%
16Claude Fable 5high· Epoch AI Benchmarking Hub data61.2%
17Gemini 2.5 Pro Preview 03-25· Epoch AI Benchmarking Hub data61.2%
18Gemini 3.1 Pro Previewhigh· Epoch AI Benchmarking Hub data61.2%
19GPT-5.6 Solmax· Epoch AI Benchmarking Hub data61%
20GPT 5· Epoch AI Benchmarking Hub data60.8%
21GPT 5.5· Epoch AI Benchmarking Hub data60.8%
22Gemini 3.1 Pro Preview· Epoch AI Benchmarking Hub data60.7%
23Grok 4.20reasoning, 0309· Epoch AI Benchmarking Hub data60.6%
24Kimi K3max· Epoch AI Benchmarking Hub data60.6%
25Gemini 3 Pro Preview· Epoch AI Benchmarking Hub data60.5%
26Gemini 3.5 Flash· Epoch AI Benchmarking Hub data60.5%
27o4 Mini· Epoch AI Benchmarking Hub data60.5%
28gemini-2.5-flash-preview-04-17· Epoch AI Benchmarking Hub data60.4%
29GLM 5· Epoch AI Benchmarking Hub data60.3%
30Gemini 3 Flash· Epoch AI Benchmarking Hub data60.2%
31GPT-6 Lunahigh· Epoch AI Benchmarking Hub data60.2%
32Grok 4.20· Epoch AI Benchmarking Hub data60.2%
33OpenAI o3· Epoch AI Benchmarking Hub data60.2%
34Claude Opus 4.1· Epoch AI Benchmarking Hub data60.1%
35Claude Sonnet 4.616k· Epoch AI Benchmarking Hub data60.1%
36Grok 4.1 Fastreasoning· Epoch AI Benchmarking Hub data60.1%
37Claude Opus 4.7· Epoch AI Benchmarking Hub data60%
38Claude Sonnet 4.6· Epoch AI Benchmarking Hub data60%
39Grok 4 Fast· Epoch AI Benchmarking Hub data60%
40Claude Sonnet 516k· Epoch AI Benchmarking Hub data59.9%
41GPT-5 mini· Epoch AI Benchmarking Hub data59.9%
42Grok 4· Epoch AI Benchmarking Hub data59.9%
43Grok 4.3· Epoch AI Benchmarking Hub data59.9%
44Claude Opus 4.5· Epoch AI Benchmarking Hub data59.8%
45Claude Opus 4.6· Epoch AI Benchmarking Hub data59.8%
46Claude Sonnet 4.5· Epoch AI Benchmarking Hub data59.8%
47GPT 5.2· Epoch AI Benchmarking Hub data59.8%
48MiniMax M3· Epoch AI Benchmarking Hub data59.8%
49Gemini 2.5 Pro· Epoch AI Benchmarking Hub data59.7%
50GPT-4.5 Preview· Epoch AI Benchmarking Hub data59.7%

Showing 1–50 of 95 results.

Data credits: Epoch AI Benchmarking Hub, CC BY 4.0 (licence)