List your company

Question Answering on SealQA 111 questions

49.5AccuracySkillSmith
2.714.852739.15May 31, 2026

Evaluation results

MethodLinks
SkillSmithBackbone=Qwen3.5-397B2026.0549.555
EvoSkillBackbone=Qwen3.5-397B2026.0541.446
Base AgentBackbone=Qwen3.5-397B2026.0530.634
SkillSmithBackbone=Qwen3.5-122B2026.0527.931
EvoSkillBackbone=Qwen3.5-122B2026.0523.426
Base AgentBackbone=Qwen3.5-122B2026.0517.119
SkillSmithBackbone=Qwen3.5-35B2026.0517.119
EvoSkillBackbone=Qwen3.5-35B2026.0514.416
Base AgentBackbone=Qwen3.5-35B2026.0510.812
SkillSmithBackbone=Qwen3.5-9B2026.056.37
EvoSkillBackbone=Qwen3.5-9B2026.055.46
Base AgentBackbone=Qwen3.5-9B2026.054.55