MDL
Qwen2.5-Math-1.5B
Qwen
Parameters
1.5B
Downloads
756.7K
on HF
Likes
109
Avg score
—
Weights
Open
Released
Sep 16, 2024
About
Qwen2.5-Math-1.5B is a 1.5B-parameter open-weights model in the Qwen2 family from Qwen.
apache-2.0text-generation
Benchmark scores · 0
No benchmark scores yet.
The research behind this model
all papers mentioning it →Qwen2.5-Math Technical Report: Toward Mathematical Expert Model via Self-ImprovementGuiding LLM Post-training Data Engineering with Model Internals from Sparse Autoencoders▲ 15 on HF · May 26, 2026You Only Need Minimal RLVR Training: Extrapolating LLMs via Rank-1 TrajectoriesMay 20, 2026Weak-to-Strong Elicitation via Mismatched Wrong DraftsMay 17, 2026CoDistill-GRPO: A Co-Distillation Recipe for Efficient Group Relative Policy OptimizationMay 9, 2026Adaptive Negative Reinforcement for LLM Reasoning:Dynamically Balancing Correction and Diversity in RLVRMay 8, 2026
Mentions matched by name in title/abstract — from the arXiv + HF daily corpus.