Claw-Eval Pass^3

Claw-Eval Pass^3 - third-pass pass rate on the Claw-Eval agentic coding benchmark.

Category

Agentic Coding

Max Score

100.0

Score Type

percent

Active

Yes

Model Scores

Scores for Claw-Eval Pass^3

Model Score Date Verified
Qwen3.6-35B-A3B
55.07%
16.04.2026 Verified
Gemma4-31B-it
25.00
16.04.2026 Unverified
Qwen3.5-35B-A3B
57.27%
16.04.2026 Unverified
Gemma4-26B-A4B
6.61%
16.04.2026 Unverified
Qwen3.6-27B
78.41%
22.04.2026 Verified
Qwen3.5-397B-A17B
50.88%
22.04.2026 Unverified
Claude 4.5 Opus
76.21%
22.04.2026 Unverified
Qwen3.5-27B
46.70%
16.04.2026 Unverified
Qwen3.8-27B
71.37%
18.08.2026 Verified
Qwen3.7-Plus
71.37%
18.08.2026 Unverified
Muse Glimmer-30B
60.57%
18.08.2026 Unverified
Opus4.6 Max
60.57%
18.08.2026 Unverified
Kimi K2.6
82.16%
23.08.2026 Verified
GPT-5.4
77.75%
23.08.2026 Verified
Claude Opus 4.6
100.00%
23.08.2026 Verified
Gemini 3.1 Pro
72.25%
23.08.2026 Verified
Kimi K2.5
60.13%
23.08.2026 Verified
Qwen3.8-Flash-Next
86.78%
26.08.2026 Verified
Claude-Opus-4.6 (Max)
60.57%
26.08.2026 Unverified