Skip to content

Benchmark

Creative Writing v3

Models measured
135
Scores recorded
138
Sources
2
IQ weight
1.57

Score distribution

1,302.9
Mean
1,436.1
Median
572.4
Std dev
450.3
10th percentile
1,914.9
90th percentile
0.803
Min
0.8032,176.9
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.57 for chat models, measured over 135 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 Astra· EQ-Bench boards2,176.9
2Kimi K3.1 Pro· EQ-Bench boards2,172.6
3Claude Fable 5.1· EQ-Bench boards2,155.3
4Claude Opus 5· EQ-Bench boards2,135
5GPT-6 Sol· EQ-Bench boards2,126
6Kimi K3· EQ-Bench boards2,083.3
7GLM 5.3· EQ-Bench boards2,076.4
8Claude Opus 5.5· EQ-Bench boards2,050.9
9Grok 4.7· EQ-Bench boards2,012.2
10Muse Wonder· EQ-Bench boards2,007
11GPT-5.6 Sol· EQ-Bench boards1,973.5
12Claude Fable 5· EQ-Bench boards1,943.6
13Aion 3.5· EQ-Bench boards1,930.6
14Muse Spark 1.1· EQ-Bench boards1,928.2
15Claude Opus 4.7· EQ-Bench boards1,914.9
16Muse Spark 1.3· EQ-Bench boards1,906.9
17GPT-5.6 Terra· EQ-Bench boards1,856.2
18GPT 5.5· EQ-Bench boards1,844.5
19Qwen3.8 Max· EQ-Bench boards1,843.8
20Muse Spark 1.2· EQ-Bench boards1,842.2
21GPT 5.4· EQ-Bench boards1,840.8
22Claude Opus 4.8· EQ-Bench boards1,840.1
23GPT-5.6 Luna· EQ-Bench boards1,831.5
24Claude Sonnet 4.6· EQ-Bench boards1,811
25Claude Opus 4.6· EQ-Bench boards1,810.3
26Muse Glimmer 30B· EQ-Bench boards1,804.3
27Claude Sonnet 5· EQ-Bench boards1,797.9
28GLM 5.2· EQ-Bench boards1,759.2
29Gemini 4 Argon· EQ-Bench boards1,757.8
30Gemini 3.8 Flash· EQ-Bench boards1,744.4
31Kimi K2.6· EQ-Bench boards1,721
32Gemini 3.7 Flash· EQ-Bench boards1,719.5
33GPT 5.2· EQ-Bench boards1,703
34Nemotron 3 Ultra· EQ-Bench boards1,691.7
35GPT-5.3 Chat· EQ-Bench boards1,689.5
36Claude Opus 4.5· EQ-Bench boards1,686.4
37Claude Sonnet 4.5· EQ-Bench boards1,677.1
38OpenAI o3· EQ-Bench boards1,675.2
39Qwen 3.8 27B· EQ-Bench boards1,670.5
40Kimi K2 Instruct· EQ-Bench boards1,665.5
41GPT-5.4 mini· EQ-Bench boards1,664.5
42Kimi K2 Thinking· EQ-Bench boards1,630.6
43GPT 5· EQ-Bench boards1,626.4
44Inkling FP4· EQ-Bench boards1,609.1
45Gemini 3.6 Flash· EQ-Bench boards1,602.4
46GLM 5· EQ-Bench boards1,600.7
47GLM 5.1· EQ-Bench boards1,593.9
48Kimi K2.5· EQ-Bench boards1,578.8
49Claude Opus 4· EQ-Bench boards1,578
50Grok 4.5· EQ-Bench boards1,575.7

Showing 1–50 of 138 results.

Data credits: EQ-Bench (Samuel J. Paech) — cite paech2023eqbench · Data by LLM Stats