Elliott/LUFFY-Qwen-Math-1.5B-Zero
TEXT GENERATIONConcurrency Cost:1Model Size:1.5BQuant:BF16Ctx Length:32kPublished:Apr 23, 2025License:mitArchitecture:Transformer Open Weights Warm

LUFFY-Qwen-Math-1.5B-Zero is a 1.5 billion parameter model developed by Elliott, built upon the Qwen2.5-Math-1.5B-Base architecture. It utilizes the LUFFY reinforcement learning framework, which integrates off-policy reasoning traces and policy shaping to enhance learning. This model is specifically optimized for mathematical reasoning tasks, demonstrating superior performance on various math benchmarks like AIME, AMC, and MATH-500 compared to its base and instruct counterparts.

Loading preview...