List your company

General Reasoning on GSM8K, MATH-500, GPQA-Diamond, and TheoremQA Aggregate

65.78Average Pass@1 AccuracyTYLER Stage 2
40.570447.115253.6660.2048Jun 15, 2026

Evaluation results

MethodLinks
TYLER Stage 2Backbone=Qwen3-4B2026.0665.78
TYLER Stage 1Backbone=Qwen3-4B2026.0663.38
GRPOBackbone=Qwen3-4B2026.0662.65
MemGenBackbone=Qwen3-4B2026.0661.28
SoftCoTBackbone=Qwen3-4B2026.0658.56
CoTBackbone=Qwen3-4B2026.0656.56
Soft-ThinkingBackbone=Qwen3-4B2026.0656.12
TYLER Stage 2Backbone=SmolLM3-3B2026.0656.03
SwiReasoningBackbone=Qwen3-4B2026.0655.72
SFTBackbone=Qwen3-4B2026.0654.27
TYLER Stage 1Backbone=SmolLM3-3B2026.0653.43
MemGenBackbone=SmolLM3-3B2026.0651.73
SoftCoTBackbone=SmolLM3-3B2026.0650.58
GRPOBackbone=SmolLM3-3B2026.0650.43
SFTBackbone=SmolLM3-3B2026.0649.16
Soft-ThinkingBackbone=SmolLM3-3B2026.0646.98
SwiReasoningBackbone=SmolLM3-3B2026.0645.92
CoTBackbone=SmolLM3-3B2026.0641.54