Collection of LLMs continually post-trained via offline GRPO to enhance mathematical reasoning capabilities.
AI & ML interests
None defined yet.
models
6
KRAFTON/AceReason-Nemotron-1.1-Offline-GRPO-7B
8B
•
Updated
•
12
•
3
KRAFTON/OpenThinker2-Offline-GRPO-7B
8B
•
Updated
•
6
•
3
KRAFTON/OpenThinker3-Offline-GRPO-7B
8B
•
Updated
•
14
•
5
KRAFTON/KORani-v1-13B
Text Generation
•
Updated
•
48
•
7
KRAFTON/KORani-v2-13B
Text Generation
•
Updated
•
7
•
4
KRAFTON/KORani-v3-13B
Text Generation
•
Updated
•
402
•
22
datasets
0
None public yet