SmallThinker-3B-Preview is a 3.1 billion parameter language model fine-tuned from Qwen2.5-3B-Instruct by PowerInfer, specifically optimized for enhanced mathematical and reasoning capabilities. It features a 32768-token context length and demonstrates significant performance improvements on benchmarks like AIME24, GAOKAO2024, and MMLU_STEM. This model is primarily designed for efficient edge deployment on resource-constrained devices and can serve as a fast draft model for larger LLMs.
No reviews yet. Be the first to review!