Skip to content

Benchmark

Forecastbench Overall score

Scores are percentages.

Models measured
85
Scores recorded
95
Sources
1
IQ weight
1.84

Score distribution

62.5%
Mean
63.1%
Median
3.2%
Std dev
58.5%
10th percentile
65.7%
90th percentile
46%
Min
46%66.6%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.84 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
1Claude Opus 5.5high· Epoch AI Benchmarking Hub data66.6%
2GPT 5.5high· Epoch AI Benchmarking Hub data66.6%
3GPT-6 Astrahigh· Epoch AI Benchmarking Hub data66.6%
4Claude Opus 5max· Epoch AI Benchmarking Hub data66.4%
5GPT-6 Solhigh· Epoch AI Benchmarking Hub data66.3%
6Claude Opus 4.8max· Epoch AI Benchmarking Hub data66%
7Claude Fable 5.1medium· Epoch AI Benchmarking Hub data65.8%
8GPT-5.6 Solpro / max· Epoch AI Benchmarking Hub data65.8%
9Muse Spark 1.3max· Epoch AI Benchmarking Hub data65.8%
10Gemini 3.7 Flashhigh· Epoch AI Benchmarking Hub data65.7%
11GPT-5.6 Solmax· Epoch AI Benchmarking Hub data65.7%
12GPT-5.6 Solmedium· Epoch AI Benchmarking Hub data65.7%
13Claude Fable 5max· Epoch AI Benchmarking Hub data65.6%
14Gemini 3.6 Flashmedium· Epoch AI Benchmarking Hub data65.6%
15Grok 4.5high· Epoch AI Benchmarking Hub data65.6%
16Grok 4.6xhigh· Epoch AI Benchmarking Hub data65.5%
17Claude Fable 5high· Epoch AI Benchmarking Hub data65.4%
18Gemini 3.1 Pro Previewhigh· Epoch AI Benchmarking Hub data65.1%
19GPT 5· Epoch AI Benchmarking Hub data64.9%
20Gemini 3 Pro Preview· Epoch AI Benchmarking Hub data64.8%
21Claude Opus 4.5· Epoch AI Benchmarking Hub data64.5%
22Grok 4· Epoch AI Benchmarking Hub data64.5%
23Grok 4 Fast· Epoch AI Benchmarking Hub data64.5%
24GPT-4.5 Preview· Epoch AI Benchmarking Hub data64.4%
25Grok 4.1 Fastreasoning· Epoch AI Benchmarking Hub data64.4%
26Claude Opus 4.1· Epoch AI Benchmarking Hub data64.3%
27Gemini 3 Flash· Epoch AI Benchmarking Hub data64.3%
28GPT-6 Lunahigh· Epoch AI Benchmarking Hub data64.3%
29OpenAI o3· Epoch AI Benchmarking Hub data64.2%
30Gemini 2.5 Pro Preview 03-25· Epoch AI Benchmarking Hub data64%
31GPT 5.5· Epoch AI Benchmarking Hub data64%
32Grok 4.20reasoning, 0309· Epoch AI Benchmarking Hub data64%
33Gemini 3.1 Pro Preview· Epoch AI Benchmarking Hub data63.9%
34Kimi K3max· Epoch AI Benchmarking Hub data63.9%
35o4 Mini· Epoch AI Benchmarking Hub data63.9%
36Claude Sonnet 4· Epoch AI Benchmarking Hub data63.8%
37Claude Sonnet 4.6· Epoch AI Benchmarking Hub data63.8%
38Gemini 2.5 Pro· Epoch AI Benchmarking Hub data63.8%
39Gemini 3.5 Flash· Epoch AI Benchmarking Hub data63.8%
40MiniMax M3· Epoch AI Benchmarking Hub data63.8%
41Claude Sonnet 4.5· Epoch AI Benchmarking Hub data63.7%
42Claude Sonnet 4.616k· Epoch AI Benchmarking Hub data63.7%
43GPT 5.2· Epoch AI Benchmarking Hub data63.7%
44Qwen 3 235B A22B· Epoch AI Benchmarking Hub data63.7%
45Claude Sonnet 3.7· Epoch AI Benchmarking Hub data63.5%
46o3 Mini· Epoch AI Benchmarking Hub data63.4%
47GLM 5· Epoch AI Benchmarking Hub data63.3%
48Claude Opus 4.6· Epoch AI Benchmarking Hub data63.1%
49Grok 4.3· Epoch AI Benchmarking Hub data63.1%
50Claude Haiku 4.5· Epoch AI Benchmarking Hub data63%

Showing 1–50 of 95 results.

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