List your company

Multi-hop Reasoning and Fact-checking on FRAMES

90.6Average @3Tongyi-DeepResearch-30B
56.865.57574.3583.125Nov 14, 2025

Evaluation results

MethodLinks
Tongyi-DeepResearch-30BType=Research Agents2025.1190.6
MiroThinker-v1.0-72BParameters=72B2025.1187.1
MiroThinker-v1.0-30BParameters=30B2025.1185.4
Claude-4.5-SonnetType=Foundation Models with Tools2025.1185
OpenAI-o3Type=Foundation Models with Tools2025.1184
DeepSeek-V3.1Type=Foundation Models with Tools2025.1183.7
SFR-DeepResearch-20BType=Research Agents2025.1182.8
Claude-4-SonnetType=Foundation Models with Tools2025.1180.7
MiroThinker-v1.0-8BParameters=8B2025.1180.6
DeepSeek-V3.2Type=Foundation Models with Tools2025.1180.2
Kimi-ResearcherType=Research Agents2025.1178.8
WebExplorer-8B-RLType=Research Agents2025.1175.7
Kimi-K2-0905Type=Foundation Models with Tools2025.1158.1