List your company

Mathematical Reasoning on AIME24 (EM)

93.33Exact Match (EM)DeepSeek-R1-0528
86.403688.20189091.7982Nov 5, 2025

Evaluation results

MethodLinks
DeepSeek-R1-0528Maximum sequence length=128k, KV cache length=Full2025.1193.33
SnapStreamMaximum sequence length=128k, KV cache length=32k2025.1190
DeepSeek-R1-0528Maximum sequence length=32k, KV cache length=32k2025.1186.67