Web Browsing Reasoning on BrowseComp+
11Avg@8 AccuracyMAS-Orchestra
Evaluation Results
| Method | Links | |
|---|---|---|
| MAS-OrchestraOrchestration Type=Training-time Orchestration, Orchestrator LLM=Qwen-7b, Sub-agent Backbone=GPT-120b (low)2026.01 | 11 | |
| SearchAgentOrchestration Type=Standalone Agent, Sub-agent Backbone=GPT-120b (low)2026.01 | 8.56 | |
| ToolOrchestraOrchestration Type=Public Training-time Orchestration, Sub-agent Backbone=GPT-120b (low)2026.01 | 1.38 | |
| CoTAgentOrchestration Type=Standalone Agent, Sub-agent Backbone=GPT-120b (low)2026.01 | 1.12 | |
| ReflexionAgentOrchestration Type=Standalone Agent, Sub-agent Backbone=GPT-120b (low)2026.01 | 1 | |
| DebateAgentOrchestration Type=Standalone Agent, Sub-agent Backbone=GPT-120b (low)2026.01 | 0.81 | |
| SCAgentOrchestration Type=Standalone Agent, Sub-agent Backbone=GPT-120b (low)2026.01 | 0.75 |