Skip to content

Benchmark

Claw-Eval v1.1 · Multimodal (101) · Pass@3

Scores are percentages. Higher is better.

Models and systems measured
11
Scores recorded
11
Sources
1
IQ weight
1.36

Score distribution

41.4%
Mean
38.6%
Median
6.7%
Std dev
34.6%
10th percentile
52.5%
90th percentile
34.6%
Min
34.6%55.5%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.36 for chat models, measured over 10 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#Model or systemScore
1GPT 5.4Claw-Eval v1.1 · multimodal · 101 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs55.5%
2Claude Opus 4.6Claw-Eval v1.1 · multimodal · 101 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs52.5%
3Kimi K2.6Claw-Eval v1.1 · multimodal · 101 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs44.6%
4Claude Sonnet 4.6Claw-Eval v1.1 · multimodal · 101 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs43.6%
5Gemini 3.1 Pro PreviewClaw-Eval v1.1 · multimodal · 101 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs39.6%
6MiMo V2.5Claw-Eval v1.1 · multimodal · 101 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs38.6%
7Gemini 3 FlashClaw-Eval v1.1 · multimodal · 101 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs37.6%
8Qwen3.5 397B A17BClaw-Eval v1.1 · multimodal · 101 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs37.6%
9Kimi K2.5Claw-Eval v1.1 · multimodal · 101 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs36.6%
10MiMo V2 OmniClaw-Eval v1.1 · multimodal · 101 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs34.6%

Not ranked

The source marks these results as descriptive, not capability scores, so we list them by name instead of ranking them.

Model or systemScore
GLM 5V TurboClaw-Eval v1.1 · multimodal · 101 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs34.6%

Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links.