List your company
ResearchFormerly WizWand

General AI Assistant Tasks on GAIA (avg@8)

88.5Avg@8 ScoreMiroThinker-H1
48.4658.85569.2579.645Nov 14, 2025Dec 4, 2025Dec 24, 2025Jan 14, 2026Feb 3, 2026Feb 23, 2026Mar 16, 2026
Updated 6mo ago

Evaluation results

MethodLinks
MiroThinker-H1Agent Style=ReAct, Max interaction turns=200, Judge LLM=gpt-4.1-2025-04-14, Max episode retries=5, Context management retention=52026.0388.5
MiroThinker-1.7Agent Style=ReAct, Max interaction turns=200, Judge LLM=gpt-4.1-2025-04-14, Max episode retries=5, Context management retention=52026.0382.7
MiroThinker-v1.0-72BParameters=72B2025.1181.9
MiroThinker-1.7-miniAgent Style=ReAct, Max interaction turns=200, Judge LLM=gpt-4.1-2025-04-14, Max episode retries=5, Context management retention=52026.0380.3
OpenAI-GPT-5Agent Style=ReAct, Max interaction turns=200, Judge LLM=gpt-4.1-2025-04-14, Max episode retries=5, Context management retention=52026.0376.4
OpenAI-GPT-5-highType=Foundation Models with Tools2025.1176.4
Minimax-M2Type=Foundation Models with Tools2025.1175.7
MiroThinker-v1.0-30BParameters=30B2025.1173.5
GLM-4.6Type=Foundation Models with Tools2025.1171.9
Claude-4.5-SonnetType=Foundation Models with Tools2025.1171.2
Tongyi-DeepResearch-30BAgent Style=ReAct, Max interaction turns=200, Judge LLM=gpt-4.1-2025-04-14, Max episode retries=5, Context management retention=52026.0370.9
Tongyi-DeepResearch-30BType=Research Agents2025.1170.9
Claude-4-SonnetType=Foundation Models with Tools2025.1168.3
OpenAI DeepResearchType=Research Agents2025.1167.4
MiroThinker-v1.0-8BParameters=8B2025.1166.4
SFR-DeepResearch-20BType=Research Agents2025.1166
DeepSeek-V3.2Type=Foundation Models with Tools2025.1163.5
DeepSeek-V3.1Type=Foundation Models with Tools2025.1163.1
Kimi-K2-0905Type=Foundation Models with Tools2025.1160.2
DeepMiner-32B-RLType=Research Agents2025.1158.7
AFM-32B-RLType=Research Agents2025.1155.3
WebExplorer-8B-RLType=Research Agents2025.1150