List your company
ResearchFormerly WizWand

General Reasoning on General Reasoning Suite Average

78.3Pass@1GPT-5.4-xhigh
8.339226.502144.66562.8279Apr 27, 2026May 1, 2026May 5, 2026May 9, 2026May 13, 2026May 17, 2026May 21, 2026
Updated 5mo ago

Evaluation results

MethodLinks
GPT-5.4-xhighSize=—, # Reas. Tokens=4821.8, Reasoning Paradigm=Pretrained LLMs + Tools2026.0578.3——
MiroThinker-v1.5Size=30B, # Reas. Tokens=11295.2, Reasoning Paradigm=Unregulated Deliberation2026.0574.2——
DeepSeek-V3.2Size=685B, # Reas. Tokens=3011.5, Reasoning Paradigm=Pretrained LLMs + Tools2026.0573.2——
SR2AM-v1.0-30BSize=30B, # Reas. Tokens=5517.7, Reasoning Paradigm=Self-Regulated Simulative Reasoning2026.0571.3——
Kimi-K2.5Size=1024B, # Reas. Tokens=6413.1, Reasoning Paradigm=Pretrained LLMs + Tools2026.0570.9——
FP16Model=Qwen3-8B, Quantization Bits=FP162026.0569.81——
LAQuantModel=Qwen3-8B, Quantization Bits=4-bit, Group Size=128, Speedup=2.45×2026.0568.62——
GPT-5.4-xhighSize=—, # Reas. Tokens=8810.3, Reasoning Paradigm=Reasoning LLMs2026.0568.4——
FP16Model=Qwen3-4B, Quantization Bits=FP162026.0566.4——
LAQuantModel=Qwen3-4B, Quantization Bits=4-bit, Group Size=128, Speedup=2.0×2026.0564.81——
LAQuantModel=Qwen3-8B, Quantization Bits=3-bit, Group Size=128, Speedup=5.15×2026.0564.28——
LAQuantModel=Qwen3-8B, Quantization Bits=3-bit2026.0564.28——
DeepSeek-V3.2Size=685B, # Reas. Tokens=9306.2, Reasoning Paradigm=Reasoning LLMs2026.0562.8——
ParoQ++Model=Qwen3-8B, Quantization Bits=3-bit2026.0562.1——
GLM-4.6Size=357B, # Reas. Tokens=3580.4, Reasoning Paradigm=Pretrained LLMs + Tools2026.0560.7——
Tongyi-DeepResearchSize=30B, # Reas. Tokens=7431.8, Reasoning Paradigm=Unregulated Deliberation2026.0560.6——
GPT-OSS-120B-highSize=120B, # Reas. Tokens=2700.4, Reasoning Paradigm=Pretrained LLMs + Tools2026.0560.3——
ParoQModel=Qwen3-8B, Quantization Bits=3-bit2026.0559.48——
ASearcher-Web-QWQ-v2Size=32B, # Reas. Tokens=116752.9, Reasoning Paradigm=Unregulated Deliberation2026.0559.2——
LAQuantModel=Qwen3-4B, Quantization Bits=3-bit, Group Size=128, Speedup=3.42×2026.0558.12——
LAQuantModel=Qwen3-4B, Quantization Bits=3-bit2026.0558.12——
ParoQ++Model=Qwen3-4B, Quantization Bits=3-bit2026.0557.2——
Qwen3-235BSize=235B, # Reas. Tokens=6467.4, Reasoning Paradigm=Pretrained LLMs + Tools2026.0557——
SR2AM-v0.1-8BSize=8B, # Reas. Tokens=3697.6, Reasoning Paradigm=Self-Regulated Simulative Reasoning2026.0557——
R-QATModel=Qwen3-8B, Quantization Bits=3-bit2026.0556——
K2-Think-V2-highSize=73B, # Reas. Tokens=31905.2, Reasoning Paradigm=Reasoning LLMs2026.0555.3——
WebExplorer-8BSize=8B, # Reas. Tokens=3616.9, Reasoning Paradigm=Unregulated Deliberation2026.0554.7——
Qwen3-30B-A3BSize=30B, # Reas. Tokens=5410.1, Reasoning Paradigm=Pretrained LLMs + Tools2026.0553.1——
Qwen3-30B-A3BSize=30B, # Reas. Tokens=8269.1, Reasoning Paradigm=Reasoning LLMs2026.0552.6——
WebSailor-32BSize=32B, # Reas. Tokens=1055.2, Reasoning Paradigm=Unregulated Deliberation2026.0551.8——
A2FMSize=32B, # Reas. Tokens=23424.8, Reasoning Paradigm=Mode Routing2026.0551.4——
GPTQModel=Qwen3-8B, Quantization Bits=3-bit2026.0551.15——
ParoQModel=Qwen3-4B, Quantization Bits=3-bit2026.0549.33——
R-QATModel=Qwen3-4B, Quantization Bits=3-bit2026.0548.91——
FP16Model=R1-Distill-Llama-8B, Quantization Bits=FP162026.0546.29——
FP16Model=Qwen3-1.7B, Quantization Bits=FP162026.0546.27——
LAQuantModel=R1-Distill-Llama-8B, Quantization Bits=4-bit, Group Size=128, Speedup=2.46×2026.0544.68——
LAQuantModel=Qwen3-1.7B, Quantization Bits=4-bit, Group Size=128, Speedup=1.64×2026.0543.8——
LAQuantModel=R1-Distill-Llama-8B, Quantization Bits=3-bit, Group Size=128, Speedup=5.56×2026.0539.54——
LAQuantModel=R1-Distill-Llama-8B, Quantization Bits=3-bit2026.0539.54——
SimpleTIR-32BSize=32B, # Reas. Tokens=5174.8, Reasoning Paradigm=Unregulated Deliberation2026.0538.6——
SASmax context length=4K, ratio r=0.32026.0438.32,7290.45
R-QATModel=R1-Distill-Llama-8B, Quantization Bits=3-bit2026.0537.8——
LAPO-Imax context length=4K2026.0437.773,3310.27
ParoQ++Model=R1-Distill-Llama-8B, Quantization Bits=3-bit2026.0537.73——
ThinkPrune-4kmax context length=4K2026.0437.623,1270.31
DeepScaleRmax context length=4K2026.0437.444,4160
L1-Maxmax context length=4K2026.0437.222,2420.46
LAQuantModel=Qwen3-1.7B, Quantization Bits=3-bit, Group Size=128, Speedup=2.29×2026.0536.6——
LAQuantModel=Qwen3-1.7B, Quantization Bits=3-bit2026.0536.6——
GRPO-4Kmax context length=4K2026.0436.552,4960.32
ParoQ++Model=Qwen3-1.7B, Quantization Bits=3-bit2026.0534.54——
WebSailor-7BSize=7B, # Reas. Tokens=2211.5, Reasoning Paradigm=Unregulated Deliberation2026.0532.8——
R-QATModel=Qwen3-1.7B, Quantization Bits=3-bit2026.0532.55——
GPTQModel=R1-Distill-Llama-8B, Quantization Bits=3-bit2026.0531.13——
SimpleTIR-7BSize=7B, # Reas. Tokens=3551.7, Reasoning Paradigm=Unregulated Deliberation2026.0530.9——
AFM-Code-7B-RLSize=7B, # Reas. Tokens=11205.5, Reasoning Paradigm=Workflow Distillation2026.0528.9——
AFM-Web-7B-RLSize=7B, # Reas. Tokens=2608.6, Reasoning Paradigm=Workflow Distillation2026.0527.5——
GPTQModel=Qwen3-4B, Quantization Bits=3-bit2026.0526.01——
ParoQModel=Qwen3-1.7B, Quantization Bits=3-bit2026.0524.57——
ASearcher-Web-7BSize=7B, # Reas. Tokens=601.4, Reasoning Paradigm=Unregulated Deliberation2026.0524.5——
ParoQModel=R1-Distill-Llama-8B, Quantization Bits=3-bit2026.0518.17——
GPTQModel=Qwen3-1.7B, Quantization Bits=3-bit2026.0511.03——