List your company

Math Reasoning on GSM8K (Accuracy and Token Delta)

96.7AccuracyQwen3-Next-80B
59.88469.4427988.558Oct 1, 2025Nov 2, 2025Dec 5, 2025Jan 6, 2026Feb 8, 2026Mar 12, 2026Apr 14, 2026

Evaluation results

MethodLinks
Qwen3-Next-80BBackbone=Qwen3-Next-80B, Decoding Strategy=Base2025.1096.7—
Qwen3-32B + THINKBRAKEBackbone=Qwen3-32B, Decoding Strategy=THINKBRAKE2025.1096.5-9.1
Qwen3-Next-80B + THINKBRAKEBackbone=Qwen3-Next-80B, Decoding Strategy=THINKBRAKE2025.1096.5-24.3
Qwen3-14BBackbone=Qwen3-14B, Decoding Strategy=Base2025.1096—
Qwen3-32BBackbone=Qwen3-32B, Decoding Strategy=Base2025.1096—
Qwen3-14B + NoWaitBackbone=Qwen3-14B, Decoding Strategy=NoWait2025.1095.8-29.4
Qwen3-32B + NoWaitBackbone=Qwen3-32B, Decoding Strategy=NoWait2025.1095.3-23.6
Qwen3-4B-Thinking + Dynasor-CoTBackbone=Qwen3-4B-Thinking, Decoding Strategy=Dynasor-CoT2025.1095.2-34
Qwen3-4B-ThinkingBackbone=Qwen3-4B-Thinking, Decoding Strategy=Base2025.1095.1—
Qwen3-4B-Thinking + DEERBackbone=Qwen3-4B-Thinking, Decoding Strategy=DEER2025.1095.1-34.1
Qwen3-14B + THINKBRAKE-pBackbone=Qwen3-14B, Decoding Strategy=THINKBRAKE-p2025.1095.1-31.6
Qwen3-14B + THINKBRAKEBackbone=Qwen3-14B, Decoding Strategy=THINKBRAKE2025.1095-15.4
Qwen3-4B-Thinking + THINKBRAKEBackbone=Qwen3-4B-Thinking, Decoding Strategy=THINKBRAKE2025.1094.8-18.9
Qwen3-32B + THINKBRAKE-pBackbone=Qwen3-32B, Decoding Strategy=THINKBRAKE-p2025.1094.8-23
Qwen3-4B + NoWaitBackbone=Qwen3-4B, Decoding Strategy=NoWait2025.1094.5-40.4
Qwen3-4B + THINKBRAKEBackbone=Qwen3-4B, Decoding Strategy=THINKBRAKE2025.1094.5-30.1
Qwen3-4BBackbone=Qwen3-4B, Decoding Strategy=Base2025.1094.4—
Qwen3-4B-Thinking + NoWaitBackbone=Qwen3-4B-Thinking, Decoding Strategy=NoWait2025.1094.2-25.1
Qwen3-4B-Thinking + THINKBRAKE-pBackbone=Qwen3-4B-Thinking, Decoding Strategy=THINKBRAKE-p2025.1093.9-37.3
Qwen3-14B + ThinkLessBackbone=Qwen3-14B, Decoding Strategy=ThinkLess2025.1093.4-100
Qwen3-32B + ThinkLessBackbone=Qwen3-32B, Decoding Strategy=ThinkLess2025.1093.3-100
Qwen3-4B + THINKBRAKE-pBackbone=Qwen3-4B, Decoding Strategy=THINKBRAKE-p2025.1092.8-40.2
DeepSeek-R1-7BBackbone=DeepSeek-R1-7B, Decoding Strategy=Base2025.1092.7—
Qwen3-4B-Thinking + ThinkLessBackbone=Qwen3-4B-Thinking, Decoding Strategy=ThinkLess2025.1092.6-100
Phi-4-Reasoning + ThinkLessBackbone=Phi-4-Reasoning, Decoding Strategy=ThinkLess2025.1092.6-100
DeepSeek-R1-7B + THINKBRAKEBackbone=DeepSeek-R1-7B, Decoding Strategy=THINKBRAKE2025.1092.4-25
Phi-4-Reasoning + NoWaitBackbone=Phi-4-Reasoning, Decoding Strategy=NoWait2025.1092.1-2.4
Phi-4-ReasoningBackbone=Phi-4-Reasoning, Decoding Strategy=Base2025.1091.7—
Phi-4-Reasoning + THINKBRAKEBackbone=Phi-4-Reasoning, Decoding Strategy=THINKBRAKE2025.1091.7-25.6
Phi-4-Reasoning + THINKBRAKE-pBackbone=Phi-4-Reasoning, Decoding Strategy=THINKBRAKE-p2025.1091.3-42.5
Qwen3-4B + ThinkLessBackbone=Qwen3-4B, Decoding Strategy=ThinkLess2025.1091-100
DeepSeek-R1-7B + NoWaitBackbone=DeepSeek-R1-7B, Decoding Strategy=NoWait2025.1090.6-30.8
GWO/n=6+CoTBase=Qwen2.52026.0490.6—
GWO+CoTBase=Qwen2.52026.0489.7—
GWOBase=Qwen2.52026.0489.1—
DeepSeek-R1-7B + DEERBackbone=DeepSeek-R1-7B, Decoding Strategy=DEER2025.1089-60.5
OMI2 (Full)Base=Qwen2.52026.0488.5—
DeepSeek-R1-7B + THINKBRAKE-pBackbone=DeepSeek-R1-7B, Decoding Strategy=THINKBRAKE-p2025.1087.8-54.9
WIBase=Qwen2.52026.0487—
WI (Full)Base=Qwen2.52026.0487—
OC-SFT-1Base=Qwen2.52026.0486.7—
CODEI/OBase=Qwen2.52026.0486.4—
DeepSeek-R1-7B + ThinkLessBackbone=DeepSeek-R1-7B, Decoding Strategy=ThinkLess2025.1085.9-100
PyEduBase=Qwen2.52026.0485.8—
CODEI/O++Base=Qwen2.52026.0485.7—
OMI2Base=Qwen2.52026.0484.1—
CoT-SC/n=5Base=Qwen2.52026.0480.1—
CoTBase=Qwen2.52026.0477.3—
DeepSeek-R1-7B + Dynasor-CoTBackbone=DeepSeek-R1-7B, Decoding Strategy=Dynasor-CoT2025.1061.3-87.8