List your company
ResearchFormerly WizWand

Reasoning on MATH (Success Score)

89.8Success Score (MATH)SInternal
36.13650.0686477.932Jan 5, 2026Jan 25, 2026Feb 15, 2026Mar 8, 2026Mar 28, 2026Apr 18, 2026May 9, 2026
Updated 5mo ago

Evaluation results

MethodLinks
SInternalModel Backbone=DeepSeek-R1-Distill-Qwen-7B, RL Training (GRPO)=true2026.0589.8
BaseModel Backbone=DeepSeek-R1-Distill-Qwen-14B, RL Training (GRPO)=true2026.0589.8
SafeChainModel Backbone=DeepSeek-R1-Distill-Qwen-14B, RL Training (GRPO)=true2026.0589.8
SInternalModel Backbone=DeepSeek-R1-Distill-Qwen-14B, RL Training (GRPO)=true2026.0589.6
STAR-1Model Backbone=DeepSeek-R1-Distill-Qwen-14B, RL Training (GRPO)=true2026.0589.4
STAR-1Model Backbone=DeepSeek-R1-Distill-Qwen-7B, RL Training (GRPO)=true2026.0589.2
BaseModel Backbone=DeepSeek-R1-Distill-Qwen-7B, RL Training (GRPO)=true2026.0588.8
SafeChainModel Backbone=DeepSeek-R1-Distill-Qwen-7B, RL Training (GRPO)=true2026.0586.6
SInternalModel Backbone=DeepSeek-R1-Distill-Llama-8B, RL Training (GRPO)=true2026.0584.2
STAR-1Model Backbone=DeepSeek-R1-Distill-Llama-8B, RL Training (GRPO)=true2026.0583.8
BaseModel Backbone=DeepSeek-R1-Distill-Llama-8B, RL Training (GRPO)=true2026.0581.8
SafeChainModel Backbone=DeepSeek-R1-Distill-Llama-8B, RL Training (GRPO)=true2026.0580
Qwen-2.5-7BTurkish Support=Advanced, Tokenization Efficiency=High, License Type=Apache 2.02026.0175.5
Llama-3.1-8BTurkish Support=Limited, Tokenization Efficiency=Low, License Type=Llama Community2026.0151.9
Mistral-7B-v0.3Turkish Support=Moderate, Tokenization Efficiency=Moderate, License Type=Apache 2.02026.0138.2