List your company
ResearchFormerly WizWand

Reasoning on HLE (Score)

64.7ScoreClaude Mythos Preview
-2.58814.88132.3549.819Jun 14, 2025Aug 17, 2025Oct 20, 2025Dec 24, 2025Feb 26, 2026May 1, 2026Jul 5, 2026
Updated 3mo ago

Evaluation results

MethodLinks
Claude Mythos PreviewOpen Source=false2025.0664.7
AGENTORCHESTRA EvolvedOpen Source=true2025.0659.6
GPT-5.4 ProOpen Source=false2025.0658.7
Muse Spark (Contemplating)Open Source=false2025.0658.4
GPT-5.5 ProOpen Source=false2025.0657.2
AGENTORCHESTRA VanillaOpen Source=true2025.0655.2
Poetiq Meta-SystemOpen Source=false2025.0655
Claude Opus 4.7Open Source=false2025.0654.7
Kimi K2.6 ThinkingOpen Source=true2025.0654
Gemini 3 Deep ThinkOpen Source=false2025.0653.4
Claude Opus 4.6Open Source=false2025.0653.1
Zoom Federated AIOpen Source=false2025.0653
GPT-5.5Open Source=false2025.0652.2
GPT-5.4Open Source=false2025.0652.1
Gemini 3.1 ProOpen Source=false2025.0651.4
Muse SparkOpen Source=false2025.0650.4
Kimi K2.5 ThinkingOpen Source=true2025.0650.2
GPT-5.2 ProOpen Source=false2025.0650
Claude Sonnet 4.6Open Source=false2025.0649
DeepSeek-V4-Pro-MaxOpen Source=true2025.0648.2
Yunjue AgentOpen Source=true2025.0648
MiMo-V2.5-ProOpen Source=true2025.0648
Gemini Deep ResearchOpen Source=false2025.0646.4
Gemini 3 Pro2026.0545.8
Gemini 3 ProOpen Source=false2025.0645.8
Claude Opus 4.52026.0543.2
Quest-35B2026.0537.2
GPT-52026.0535.2
Tongyi-DR2026.0532.9
OpenAI-DR2026.0526.6
Quest-30B2026.0524.6
OpenResearcherEvaluation Implementation=Authors' implementation2026.0519.6
DeepSeek v3.22025.1217.9
Confident DecodingModel=Qwen3.5-27B, Decoding Method=Confident2026.0617
Confident DecodingModel=Qwen3.5-122B-A10B, Decoding Method=Confident2026.0616.5
Last Layer DecodingModel=Qwen3.5-27B, Decoding Method=Last Layer2026.0616
Nemotron-3-Puzzle-75B-A9BModel=Nemotron-3-Puzzle-75B-A9B, Precision=FP8, Tool Use=no tools2026.0716
DeepSeek R1 05282025.1215.9
GLM-4.52025.1214.8
Last Layer DecodingModel=Qwen3.5-122B-A10B, Decoding Method=Last Layer2026.0614.7
INTELLECT-3Parameters=100B+2025.1214.6
GLM-4.5-Air2025.1213.3
GLM-4.6Reported by AA Index=true2025.1213.3
Confident DecodingModel=gpt-oss-120B, Decoding Method=Confident2026.0612.6
Confident DecodingModel=Gemma-4-31B, Decoding Method=Confident2026.0611
Last Layer DecodingModel=gpt-oss-120B, Decoding Method=Last Layer2026.0610.8
GPT-OSSParameters=120B2025.1210.6
Confident DecodingModel=Qwen3.5-35B-A3B, Decoding Method=Confident2026.069.5
Last Layer DecodingModel=Gemma-4-31B, Decoding Method=Last Layer2026.069.2
Last Layer DecodingModel=Qwen3.5-35B-A3B, Decoding Method=Last Layer2026.067.1
Confident DecodingModel=gpt-oss-20b, Decoding Method=Confident2026.066.4
Confident DecodingModel=Qwen3.5-9B-Base, Decoding Method=Confident2026.066.3
Confident DecodingModel=Qwen3.5-9B, Decoding Method=Confident2026.066.1
Ouro-2.6B-Thinking-R4Parameters=2.6B, Configuration=Thinking-R42025.105.58
Last Layer DecodingModel=gpt-oss-20b, Decoding Method=Last Layer2026.065.5
Last Layer DecodingModel=Qwen3.5-9B-Base, Decoding Method=Last Layer2026.065.4
Ouro-1.4B-Thinking-R4Parameters=1.4B, Configuration=Thinking-R42025.105.21
Qwen3-4BParameters=4B2025.105.21
Last Layer DecodingModel=Qwen3.5-9B, Decoding Method=Last Layer2026.065.2
Deepseek-Distill-Qwen-7BParameters=7B2025.105.14
Deepseek-Distill-Qwen-1.5BParameters=1.5B2025.104.2
Qwen3-1.7BParameters=1.7B2025.104.13
Qwen3-8BParameters=8B2025.102.22
Last Layer DecodingModel=Qwen3.5-0.8B, Decoding Method=Last Layer2026.060
Confident DecodingModel=Qwen3.5-0.8B, Decoding Method=Confident2026.060