Skip to content

Benchmark

Forecastbench Market score

Scores are percentages.

Models measured
85
Scores recorded
95
Sources
1
IQ weight
1.29

Score distribution

67.1%
Mean
67.6%
Median
3.2%
Std dev
63.6%
10th percentile
70.3%
90th percentile
50.9%
Min
50.9%71.3%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.29 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 5max· Epoch AI Benchmarking Hub data71.3%
2Claude Opus 5.5high· Epoch AI Benchmarking Hub data71.3%
3GPT-5.6 Solmax· Epoch AI Benchmarking Hub data71.2%
4Claude Opus 4.8max· Epoch AI Benchmarking Hub data70.9%
5GPT-5.6 Solpro / max· Epoch AI Benchmarking Hub data70.8%
6Grok 4.5high· Epoch AI Benchmarking Hub data70.6%
7Claude Fable 5max· Epoch AI Benchmarking Hub data70.4%
8GPT-6 Astrahigh· Epoch AI Benchmarking Hub data70.4%
9Claude Fable 5high· Epoch AI Benchmarking Hub data70.3%
10Claude Fable 5.1medium· Epoch AI Benchmarking Hub data70.3%
11GPT-5.6 Solmedium· Epoch AI Benchmarking Hub data70.3%
12GPT 5.5high· Epoch AI Benchmarking Hub data70.2%
13Claude Opus 4.5· Epoch AI Benchmarking Hub data70%
14Grok 4.6xhigh· Epoch AI Benchmarking Hub data70%
15Gemini 3.6 Flashmedium· Epoch AI Benchmarking Hub data69.8%
16Gemini 3.7 Flashhigh· Epoch AI Benchmarking Hub data69.8%
17GPT-4.5 Preview· Epoch AI Benchmarking Hub data69.8%
18Grok 4· Epoch AI Benchmarking Hub data69.8%
19Gemini 3 Pro Preview· Epoch AI Benchmarking Hub data69.7%
20Claude Sonnet 4· Epoch AI Benchmarking Hub data69.6%
21GPT 5· Epoch AI Benchmarking Hub data69.6%
22Grok 4 Fast· Epoch AI Benchmarking Hub data69.6%
23Muse Spark 1.3max· Epoch AI Benchmarking Hub data69.6%
24Gemini 3.1 Pro Previewhigh· Epoch AI Benchmarking Hub data69.4%
25Grok 4.1 Fastreasoning· Epoch AI Benchmarking Hub data69.3%
26GPT-6 Solhigh· Epoch AI Benchmarking Hub data69.2%
27Claude Opus 4.1· Epoch AI Benchmarking Hub data69.1%
28Gemini 3 Flash· Epoch AI Benchmarking Hub data69.1%
29GPT-6 Lunahigh· Epoch AI Benchmarking Hub data69%
30o3 Mini· Epoch AI Benchmarking Hub data68.8%
31GPT-4 Turbo· Epoch AI Benchmarking Hub data68.7%
32OpenAI o3· Epoch AI Benchmarking Hub data68.7%
33Claude Sonnet 3.7· Epoch AI Benchmarking Hub data68.6%
34Gemini 2.5 Pro· Epoch AI Benchmarking Hub data68.5%
35Qwen 3 235B A22B· Epoch AI Benchmarking Hub data68.4%
36MiniMax M3· Epoch AI Benchmarking Hub data68.3%
37GPT 5.2· Epoch AI Benchmarking Hub data68.2%
38Claude Sonnet 4.5· Epoch AI Benchmarking Hub data68.1%
39Claude Sonnet 4.6· Epoch AI Benchmarking Hub data68.1%
40DeepSeek V3· Epoch AI Benchmarking Hub data68.1%
41Claude Haiku 4.5· Epoch AI Benchmarking Hub data67.9%
42Claude 3.5 Sonnet· Epoch AI Benchmarking Hub data67.8%
43GLM 4.5 Air· Epoch AI Benchmarking Hub data67.8%
44Llama 3.3 70B Instruct· Epoch AI Benchmarking Hub data67.8%
45Mistral Large2407· Epoch AI Benchmarking Hub data67.8%
46Claude Sonnet 4.616k· Epoch AI Benchmarking Hub data67.7%
47Grok 4.20reasoning, 0309· Epoch AI Benchmarking Hub data67.7%
48Kimi K3max· Epoch AI Benchmarking Hub data67.6%
49o4 Mini· Epoch AI Benchmarking Hub data67.6%
50Gemini 3.5 Flash· Epoch AI Benchmarking Hub data67.5%

Showing 1–50 of 95 results.

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