Latest AI Insights

Model Evaluations

Model Directory

Model Comparison

Resource Center

Search blog

加载中...

DataLearner AI

A knowledge platform focused on LLM benchmarking, datasets, and practical instruction with continuously updated capability maps.

产品

Leaderboards
模型对比
Datasets

资源

Tutorials
Editorial
Tool directory

关于

关于我们
隐私政策
数据收集方法
联系我们

© 2026 DataLearner AI. DataLearner curates industry data and case studies so researchers, enterprises, and developers can rely on trustworthy intelligence.

隐私政策服务条款

MPT-7B-Base - AI Model Details and Benchmarks | DataLearnerAI

Page navigation

目录

Model catalogMPT-7B-Base

MP

MPT-7B-Base

MosaicML Pretrained Transformer - 7B Base

Release date: 2023-05-05更新于: 2023-06-23 20:19:43.390121

Live demoGitHub Hugging Face Compare

Parameters

67.0亿

Context length

2K

Chinese support

Not supported

Reasoning ability

Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology

MPT-7B-Base

Model basics

Reasoning traces

Not supported

Context length

2K tokens

Max output length

No data

Model type

基础大模型

Release date

2023-05-05

Model file size

MPT-7B-Base

Open source & experience

Code license

Weights license

Apache 2.0- 免费商用授权

GitHub repo

https://github.com/mosaicml/llm-foundry

Hugging Face

https://huggingface.co/mosaicml/mpt-7b

MPT-7B-Base

Official resources

Paper

Introducing MPT-7B: A New Standard for Open-Source, Commercially Usable LLMs

DataLearnerAI blog

No blog post yet

MPT-7B-Base

API details

API speed

No data

No public API pricing yet.

MPT-7B-Base

Benchmark Results

No benchmark data to show.

MPT-7B-Base

Publisher

MosaicML

View publisher details

MosaicML Pretrained Transformer - 7B Base

Model Overview

MPT-7B是由MosaicML推出的transformer系列大模型，是基于1万亿tokens的文本和代码训练的。这是一个完全开源且允许商用的大模型，质量与LLaMA-7B差不多。

MPT系列介绍： https://www.datalearner.com/ai-models/foundation-models/MPT

MPT-7B-Base是其中的基础模型，是一个decoder-style transformer，参数为67亿。它是在MosaicML的数据团队收集的1万亿文本和代码数据集上进行训练的。这个基本模型使用了用于快速训练和推理的FlashAttention，以及用于微调和外推到长上下文长度的ALiBi。

Foundation model

MPT

DataLearner 官方微信

欢迎关注 DataLearner 官方微信，获得最新 AI 技术推送

DataLearner 官方微信二维码

13.3GB

MoE architecture

No

Total params / Active params

67.0B / N/A

Knowledge cutoff

No data

Inference modes

No mode data

Live demo

No live demo