List your company

Question Answering on DocBench

85.1AccuracyEvolved Agent
56.29263.77171.2578.729Jun 28, 2026

Evaluation results

MethodLinks
Evolved AgentModel=Gemini-3.1-Flash, Evolved=true2026.0685.1
Evolved Agent, No Ground TruthModel=Gemini-3.1-Flash, Evolved=true2026.0684
SimpleDocModel=Gemini-3.1-Flash, Evolved=false2026.0682.2
Evolved Agent, GPT-5-minimal Meta AgentModel=Gemini-3.1-Flash, Evolved=true2026.0679.8
MACTModel=Gemini-3.1-Flash, Evolved=false2026.0679.7
Evolved AgentModel=GPT-5-mini, Evolved=true2026.0679.3
MDocAgentModel=Gemini-3.1-Flash, Evolved=false2026.0678.6
Evolved Agent, No ExplorationModel=Gemini-3.1-Flash, Evolved=true2026.0677.3
Evolved Agent, No Ground TruthModel=GPT-5-mini, Evolved=true2026.0675.6
Evolved Agent, GPT-5-minimal Meta AgentModel=GPT-5-mini, Evolved=true2026.0675.6
SimpleDocModel=GPT-5-mini, Evolved=false2026.0674.8
Evolved Agent, No ExplorationModel=GPT-5-mini, Evolved=true2026.0674.8
Baseline AgentModel=Gemini-3.1-Flash, Evolved=false2026.0673.4
MACTModel=GPT-5-mini, Evolved=false2026.0673.3
File APIModel=GPT-5-mini, Evolved=false2026.0672.1
File APIModel=Gemini-3.1-Flash, Evolved=false2026.0671.7
MDocAgentModel=GPT-5-mini, Evolved=false2026.0670.1
Baseline AgentModel=GPT-5-mini, Evolved=false2026.0668.6
CoTModel=Gemini-3.1-Flash, Evolved=false2026.0666.1
CoTModel=GPT-5-mini, Evolved=false2026.0657.4