# Mistral AI 发布数学推理模型 Mathstral 7B

- 来源：Mistral AI：News（网页）
- 发布时间：2024-07-16 00:00
- AIHOT 分数：37
- AIHOT 链接：https://aihot.virxact.com/items/cmppdcr7e0e4fslv43ff2of1k
- 原文链接：https://mistral.ai/news/mathstral

## AI 摘要

Mistral AI 发布了专注于数学推理的 7B 参数模型 Mathstral 7B。该模型基于 Mistral 7B 构建，旨在解决需要复杂多步推理的高级数学问题。它在 MATH 和 MMLU 基准上分别达到 56.6% 和 63.47%，在同等参数规模中实现 SOTA 性能。通过增加推理时计算，其在 MATH 上的分数可借助多数投票提升至 68.37%，使用强奖励模型则可达 74.59%。该模型为指令模型，权重已托管于 HuggingFace。

## 正文

We're contributing Mathstral to the science community to bolster efforts in advanced mathematical problems requiring complex, multi-step logical reasoning. The Mathstral release is part of our broader effort to support academic projects—it was produced in the context of our collaboration with Project Numina .

Akin to Isaac Newton in his time, Mathstral stands on the shoulders of Mistral 7B and specializes in STEM subjects. It achieves state-of-the-art reasoning capacities in its size category across various industry-standard benchmarks. In particular, it achieves 56.6% on MATH and 63.47% on MMLU, with the following MMLU performance difference by subject between Mathstral 7B and Mistral 7B.

Mathstral is another example of the excellent performance/speed tradeoffs achieved when building models for specific purposes – a development philosophy we actively promote in la Plateforme, particularly with its new fine-tuning capabilities .

Mathstral can achieve significantly better results with more inference-time computation: Mathstral 7B scores 68.37% on MATH with majority voting and 74.59% with a strong reward model among 64 candidates.

Mathstral is an instructed model – use it or fine-tune it as such, referring to our documentation. Weights are hosted on HuggingFace . You can try Mathstral now with mistral-inference and adapt it with mistral-finetune .

We thank Professor Paul Bourdon for curating the GRE Math Subject Test problems used in our evaluation.

0%