CMU-AIRe/e3-1.7B
TEXT GENERATIONConcurrency Cost:1Model Size:2BQuant:BF16Ctx Length:32kPublished:Jun 11, 2025License:mitArchitecture:Transformer0.0K Open Weights Warm
CMU-AIRe/e3-1.7B is a 1.7 billion parameter language model developed by CMU-AIRe, specifically designed for mathematical reasoning. It is trained on specialized datasets for mathematical problems, including easy and medium-hard stages. With a 32768 token context length, this model is optimized for solving complex mathematical challenges.
Loading preview...
CMU-AIRe/e3-1.7B: A Specialized Math Reasoning Model
CMU-AIRe/e3-1.7B is a 1.7 billion parameter language model developed by CMU-AIRe, engineered with a strong focus on mathematical problem-solving. This model distinguishes itself through its dedicated training regimen, utilizing specialized datasets to enhance its mathematical reasoning capabilities.
Key Capabilities
- Mathematical Reasoning: Optimized for tackling mathematical problems across various difficulty levels.
- Extensive Context Window: Features a 32768 token context length, allowing it to process and understand longer mathematical problem descriptions and solution steps.
- Targeted Training: Benefits from a two-stage training process on curated mathematical datasets:
- Stage 1: Focused on foundational mathematical concepts using the e3-math-easy dataset.
- Stage 2: Advanced training on more complex problems using the e3-math-medhard dataset.
Good For
- Mathematical Problem Solving: Ideal for applications requiring accurate and robust solutions to mathematical questions.
- Educational Tools: Can be integrated into platforms for tutoring, homework assistance, or generating math problems.
- Research in AI for Math: Useful for researchers exploring advanced mathematical reasoning in language models.