List your company
ResearchFormerly WizWand

Expert-level Reasoning on Humanity's Last Exam 2,158 text-only

54.2Avg@3 ScoreSeed-2.0-Pro
32.04837.79943.5549.301Mar 16, 2026
Updated 7mo ago

Evaluation results

MethodLinks
Seed-2.0-ProAgent Style=ReAct, Max interaction turns=200, Judge LLM=o3-mini-2025-01-31, Max episode retries=5, Context management retention=52026.0354.2
Claude-4.6-OpusAgent Style=ReAct, Max interaction turns=200, Judge LLM=o3-mini-2025-01-31, Max episode retries=5, Context management retention=52026.0353.1
OpenAI-GPT-5.4Agent Style=ReAct, Max interaction turns=200, Judge LLM=o3-mini-2025-01-31, Max episode retries=5, Context management retention=52026.0352.1
Gemini-3.1-ProAgent Style=ReAct, Max interaction turns=200, Judge LLM=o3-mini-2025-01-31, Max episode retries=5, Context management retention=52026.0351.4
GLM-5.0Agent Style=ReAct, Max interaction turns=200, Judge LLM=o3-mini-2025-01-31, Max episode retries=5, Context management retention=52026.0350.4
Kimi-K2.5Agent Style=ReAct, Max interaction turns=200, Judge LLM=o3-mini-2025-01-31, Max episode retries=5, Context management retention=52026.0350.2
Qwen3.5-397BAgent Style=ReAct, Max interaction turns=200, Judge LLM=o3-mini-2025-01-31, Max episode retries=5, Context management retention=52026.0348.3
MiroThinker-H1Agent Style=ReAct, Max interaction turns=200, Judge LLM=o3-mini-2025-01-31, Max episode retries=5, Context management retention=52026.0347.7
Gemini-3.0-ProAgent Style=ReAct, Max interaction turns=200, Judge LLM=o3-mini-2025-01-31, Max episode retries=5, Context management retention=52026.0346.9
Claude-4.5-OpusAgent Style=ReAct, Max interaction turns=200, Judge LLM=o3-mini-2025-01-31, Max episode retries=5, Context management retention=52026.0343.2
MiroThinker-1.7Agent Style=ReAct, Max interaction turns=200, Judge LLM=o3-mini-2025-01-31, Max episode retries=5, Context management retention=52026.0342.9
DeepSeek-V3.2Agent Style=ReAct, Max interaction turns=200, Judge LLM=o3-mini-2025-01-31, Max episode retries=5, Context management retention=52026.0340.8
MiroThinker-1.7-miniAgent Style=ReAct, Max interaction turns=200, Judge LLM=o3-mini-2025-01-31, Max episode retries=5, Context management retention=52026.0336.4
OpenAI-GPT-5Agent Style=ReAct, Max interaction turns=200, Judge LLM=o3-mini-2025-01-31, Max episode retries=5, Context management retention=52026.0335.2
Tongyi-DeepResearch-30BAgent Style=ReAct, Max interaction turns=200, Judge LLM=o3-mini-2025-01-31, Max episode retries=5, Context management retention=52026.0332.9