List your company

Reasoning on 10 challenging reasoning tasks Combined

46.7Average ScoreE3-TIR
26.00431.37736.7542.123Apr 10, 2026

Evaluation results

MethodLinks
E3-TIRBackbone=Qwen2.5-3B, Agent Category=Multi-Tool Integrated Agent2026.0446.7
ARPOBackbone=Qwen2.5-3B, Agent Category=Multi-Tool Integrated Agent2026.0445.2
Tool-StarBackbone=Qwen2.5-3B, Agent Category=Multi-Tool Integrated Agent2026.0444.3
ReCallBackbone=Qwen2.5-3B, Agent Category=Multi-Tool Integrated Agent2026.0440.5
Tree-GRPOBackbone=Qwen2.5-3B, Agent Category=LLM-based Search Agent2026.0436.1
Search-R1Backbone=Qwen2.5-3B, Agent Category=LLM-based Search Agent2026.0435.6
ToRLBackbone=Qwen2.5-3B, Agent Category=LLM-based Math Agent2026.0432.2
SimpleTIRBackbone=Qwen2.5-3B, Agent Category=LLM-based Math Agent2026.0431.7
Search-o1Backbone=Qwen2.5-3B, Agent Category=LLM-based Search Agent2026.0430.7
Qwen2.5-3B-InstructBackbone=Qwen2.5-3B2026.0426.8