Qwen2-Math-1.5B|7B|72B
by Community
GITHUB HUGGING FACE MODELSCOPE DISCORD 🚨 This model mainly supports English. We will release bilingual (English and Chinese) math models soon. Introduction Over the past year, w
OSS
Qwen2-Math-1.5B|7B|72B
Added 1 June 2026
Overview
Qwen2-Math is a series of open-source large language models specialized for arithmetic and mathematical problem solving. Available in 1.5B, 7B, and 72B parameter variants, these models are built on the Qwen2 architecture and are designed to enhance reasoning capabilities for math tasks. Currently the models primarily support English, with bilingual English-Chinese versions in development.
Best for
Best for
Developers who need a math-focused reasoning model within a resource-constrained or open-source pipeline
Use cases
- Solving arithmetic and mathematical problems in applications
- Automated math tutoring or answer verification
- Embedding math reasoning into chatbots or educational tools
Notes
Qwen2-Math is a series of open-source large language models specialized for arithmetic and mathematical problem solving. Available in 1.5B, 7B, and 72B parameter variants, these models are built on the Qwen2 architecture and are designed to enhance reasoning capabilities for math tasks. Currently the models primarily support English, with bilingual English-Chinese versions in development.
Use cases
- Solving arithmetic and mathematical problems in applications
- Automated math tutoring or answer verification
- Embedding math reasoning into chatbots or educational tools
Pros
- Specialized for math, delivering strong performance on arithmetic reasoning
- Multiple model sizes allow trade-offs between speed and capability
- Open-source with availability on GitHub, Hugging Face, and ModelScope
Cons
- Currently limited to English; bilingual support is not yet released
- May underperform on non-mathematical language tasks compared to general models
- Larger models require significant computational resources
Indexed from awesome-llm and enriched against its public facts.
Pros
- Specialized for math, delivering strong performance on arithmetic reasoning
- Multiple model sizes allow trade-offs between speed and capability
- Open-source with availability on GitHub, Hugging Face, and ModelScope
Cons
- Currently limited to English; bilingual support is not yet released
- May underperform on non-mathematical language tasks compared to general models
- Larger models require significant computational resources
Pairs with
Other entries in the index that connect to this one. Click through to see the chain.
vLLM
Community
A high-throughput and memory-efficient inference and serving engine for LLMs
ollama
Community
Get up and running with Kimi-K2.5, GLM-5, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.
llama.cpp
Community
LLM inference in C/C++
lm-evaluation-harness
Community
A framework for few-shot evaluation of language models.
FastChat
Community
An open platform for training, serving, and evaluating large language models. Release repo for Vicuna and Chatbot Arena.