Skip to content

Benchmark

Claw-Eval v1.1 · General (161) · Pass³

Scores are percentages. Higher is better.

Models and systems measured
25
Scores recorded
25
Sources
1
IQ weight
0.23

Score distribution

55.9%
Mean
57.8%
Median
12%
Std dev
48.4%
10th percentile
68.3%
90th percentile
6.8%
Min
6.8%70.8%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.23 for chat models, measured over 20 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
1Claude Opus 4.6Claw-Eval v1.1 · general · 161 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs70.8%
2Claude Sonnet 4.6Claw-Eval v1.1 · general · 161 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs68.3%
3Step 3.7 FlashClaw-Eval v1.1 · general · 161 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs68.3%
4MiMo V2.5 ProClaw-Eval v1.1 · general · 161 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs64%
5GLM 5.1Claw-Eval v1.1 · general · 161 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs62.7%
6MiMo V2.5Claw-Eval v1.1 · general · 161 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs62.1%
7Kimi K2.6Claw-Eval v1.1 · general · 161 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs61.5%
8Agnes 2.0 FlashClaw-Eval v1.1 · general · 161 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs60.9%
9GPT 5.4Claw-Eval v1.1 · general · 161 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs60.2%
10DeepSeek V4 Pro 0813Claw-Eval v1.1 · general · 161 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs58.4%
11DeepSeek V4 FlashClaw-Eval v1.1 · general · 161 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs57.8%
12Qwen3.5 397B A17BClaw-Eval v1.1 · general · 161 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs57.8%
13MiMo V2 ProClaw-Eval v1.1 · general · 161 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs57.1%
14Qwen3.6 PlusClaw-Eval v1.1 · general · 161 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs57.1%
15Gemini 3.1 Pro PreviewClaw-Eval v1.1 · general · 161 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs55.9%
16Kimi K2.5Claw-Eval v1.1 · general · 161 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs52.8%
17MiMo V2 OmniClaw-Eval v1.1 · general · 161 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs52.2%
18MiniMax M2.7Claw-Eval v1.1 · general · 161 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs49.7%
19Gemini 3 FlashClaw-Eval v1.1 · general · 161 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs48.4%
20DeepSeek V3.2Claw-Eval v1.1 · general · 161 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs42.2%

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 5 TurboClaw-Eval v1.1 · general · 161 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs57.1%
GLM 5V TurboClaw-Eval v1.1 · general · 161 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs52.8%
Mach-Mind 4 ProClaw-Eval v1.1 · general · 161 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs50.3%
Muse Spark (version unspecified)Claw-Eval v1.1 · general · 161 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs62.7%
Nemotron 3 Super (variant unspecified)Claw-Eval v1.1 · general · 161 tasks · N=3 · grader: Gemini 3 Flash· Reported by Claw-Eval· via BenchBoard docs6.8%

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.