This, and Qwen 2.5 math and the DeepSeek Math models, are really pushing the MATH scores of open models. The DeepSeek paper is great, and supposedly more Qwen reports are coming soon.
Similarity · VAIL
VAIL Fingerprint
0072:00b9:0106:01a8:0232:03aa:058b:4aac
Explore other models with behavioral similarity to Qwen2-Math-72B-Instruct.
Adoption · Hugging Face
RAM score
—
Hugging Face Downloads
138
last 30d
31.5K
all time
HF Likes
89
Relative Adoption Metric contextualizes downloads against the model's size bucket.
Related Models