hkust-nlp/dart-math-mistral-7b-prop2diff
TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:4kPublished:Jun 5, 2024License:apache-2.0Architecture:Transformer0.0K Open Weights Cold

The hkust-nlp/dart-math-mistral-7b-prop2diff model is a 7 billion parameter Mistral-based language model developed by hkust-nlp, fine-tuned using Difficulty-Aware Rejection Tuning (DART-Math). This model is specifically optimized for mathematical problem-solving, demonstrating strong performance on various in-domain and challenging out-of-domain math benchmarks. It utilizes a 'Prop2Diff' strategy to address biases towards easier queries in training datasets, making it particularly effective for complex mathematical reasoning tasks.

Loading preview...