List your company

Web Search Question Answering on SearchQA (held-out test)

87.3Score (%)SkillOpt
54.5463.04571.5580.055May 22, 2026

Evaluation results

MethodLinks
SkillOptModel=GPT–5.5, Harness=No harness / direct chat2026.0587.3
SkillOptModel=GPT–5.5, Harness=Codex harness2026.0587.3
SkillOptModel=GPT–5.5, Harness=Claude Code harness2026.0585.9
GEPAModel=GPT–5.5, Harness=No harness / direct chat2026.0584.8
Human skillModel=GPT–5.5, Harness=Codex harness2026.0584.1
EvoSkillModel=GPT–5.5, Harness=Claude Code harness2026.0584
Human skillModel=GPT–5.5, Harness=Claude Code harness2026.0583.7
LLM skillModel=GPT–5.5, Harness=Codex harness2026.0583.4
SkillOptModel=GPT–5.4, Harness=No harness / direct chat2026.0583.1
SkillOptModel=GPT–5.2, Harness=No harness / direct chat2026.0583.1
GEPAModel=GPT–5.2, Harness=No harness / direct chat2026.0582.7
TextGradModel=GPT–5.2, Harness=No harness / direct chat2026.0582.6
TextGradModel=GPT–5.4, Harness=No harness / direct chat2026.0582.5
Trace2SkillModel=GPT–5.5, Harness=No harness / direct chat2026.0582.4
GEPAModel=GPT–5.4, Harness=No harness / direct chat2026.0582.4
LLM skillModel=GPT–5.5, Harness=Claude Code harness2026.0582.4
No skillModel=GPT–5.5, Harness=Claude Code harness2026.0581.9
Human skillModel=GPT–5.5, Harness=No harness / direct chat2026.0581.8
No skillModel=GPT–5.5, Harness=Codex harness2026.0581.8
TextGradModel=GPT–5.5, Harness=No harness / direct chat2026.0581.4
Trace2SkillModel=GPT–5.4, Harness=No harness / direct chat2026.0581.2
LLM skillModel=GPT–5.5, Harness=No harness / direct chat2026.0580.9
SkillOptModel=Qwen3.6–35B-A3B, Harness=No harness / direct chat2026.0580.3
SkillOptModel=GPT–5.4-mini, Harness=No harness / direct chat2026.0580.2
Human skillModel=GPT–5.4, Harness=No harness / direct chat2026.0580.1
LLM skillModel=GPT–5.4, Harness=No harness / direct chat2026.0579.4
GEPAModel=GPT–5.4-mini, Harness=No harness / direct chat2026.0579.4
Trace2SkillModel=GPT–5.2, Harness=No harness / direct chat2026.0579.2
Trace2SkillModel=GPT–5.4-mini, Harness=No harness / direct chat2026.0578.6
No skillModel=GPT–5.5, Harness=No harness / direct chat2026.0577.7
TextGradModel=GPT–5.4-mini, Harness=No harness / direct chat2026.0577.5
Human skillModel=GPT–5.4-mini, Harness=No harness / direct chat2026.0577.2
No skillModel=GPT–5.4, Harness=No harness / direct chat2026.0576.9
TextGradModel=Qwen3.6–35B-A3B, Harness=No harness / direct chat2026.0576.4
LLM skillModel=GPT–5.4-mini, Harness=No harness / direct chat2026.0576.1
No skillModel=GPT–5.4-mini, Harness=No harness / direct chat2026.0575.9
Human skillModel=GPT–5.2, Harness=No harness / direct chat2026.0575.8
GEPAModel=Qwen3.6–35B-A3B, Harness=No harness / direct chat2026.0575.8
Trace2SkillModel=Qwen3.6–35B-A3B, Harness=No harness / direct chat2026.0575.4
SkillOptModel=GPT–5.4-nano, Harness=No harness / direct chat2026.0574.8
Human skillModel=Qwen3.6–35B-A3B, Harness=No harness / direct chat2026.0574.1
LLM skillModel=GPT–5.2, Harness=No harness / direct chat2026.0573.7
TextGradModel=GPT–5.4-nano, Harness=No harness / direct chat2026.0573.4
GEPAModel=GPT–5.4-nano, Harness=No harness / direct chat2026.0573.2
No skillModel=Qwen3.6–35B-A3B, Harness=No harness / direct chat2026.0572.7
LLM skillModel=Qwen3.6–35B-A3B, Harness=No harness / direct chat2026.0572.6
No skillModel=GPT–5.2, Harness=No harness / direct chat2026.0571.9
SkillOptModel=Qwen3.5–4B, Harness=No harness / direct chat2026.0571.2
Trace2SkillModel=GPT–5.4-nano, Harness=No harness / direct chat2026.0569.9
GEPAModel=Qwen3.5–4B, Harness=No harness / direct chat2026.0568.6
Trace2SkillModel=Qwen3.5–4B, Harness=No harness / direct chat2026.0568.5
No skillModel=Qwen3.5–4B, Harness=No harness / direct chat2026.0568.1
Human skillModel=Qwen3.5–4B, Harness=No harness / direct chat2026.0566.3
Human skillModel=GPT–5.4-nano, Harness=No harness / direct chat2026.0566.1
LLM skillModel=Qwen3.5–4B, Harness=No harness / direct chat2026.0565
LLM skillModel=GPT–5.4-nano, Harness=No harness / direct chat2026.0562.4
EvoSkillModel=GPT–5.5, Harness=Codex harness2026.0561.4
TextGradModel=Qwen3.5–4B, Harness=No harness / direct chat2026.0560.7
No skillModel=GPT–5.4-nano, Harness=No harness / direct chat2026.0555.8