SWE Atlas - RF

SWE Atlas RF - agentic coding subtask; mini SWE agent harness, temp=1.0, top_p=0.95, 128K context, avg of 5 runs.

Category

Agentic Coding

Max Score

100.0

Score Type

percent

Active

Yes

Model Scores

Scores for SWE Atlas - RF

Model Score Date Verified
Ornith-1.0-9B
22.08%
25.06.2026 Verified
Qwen3.5-9B
4.30
25.06.2026 Unverified
Qwen3.5-35B-A3B
10.59%
25.06.2026 Unverified
Ornith-1.0-35B
45.60%
25.06.2026 Verified
Qwen3.6-35B-A3B
12.75%
25.06.2026 Unverified
Qwen3.5-397B-A17B
25.31%
25.06.2026 Unverified
Hy3
51.35%
28.08.2026 Verified
Hy4 preview
87.97%
28.08.2026 Verified
DeepSeek V4 Pro
79.53%
28.08.2026 Verified
Qwen3.8-2.4T-A95B
83.84%
28.08.2026 Verified
GLM-5.3
85.46%
28.08.2026 Verified
Kimi K3
59.43%
28.08.2026 Verified
GPT-5.6 Sol
86.36%
28.08.2026 Verified
Claude Opus 5
100.00%
28.08.2026 Verified