# Mistral AI 开放平台 Beta 访问发布

- 来源：Mistral AI：News（网页）
- 发布时间：2023-12-11 00:00
- AIHOT 分数：24
- AIHOT 链接：https://aihot.virxact.com/items/cmppdcr7f0e4pslv46kvu5y4e
- 原文链接：https://mistral.ai/news/la-plateforme

## AI 摘要

Mistral AI 开放平台提供 Beta 访问，包含三个聊天端点和一个嵌入向量端点。聊天端点分别为：提供 Mistral 7B Instruct v0.2、仅支持英语的 mistral-tiny；提供 Mixtral 8x7B、支持多语言与代码的 mistral-small；以及提供高性能原型模型的 mistral-medium。三者 MT-Bench 分数依次为 7.6、8.3 和 8.6。嵌入端点 Mistral-embed 提供 1024 维向量，在 MTEB 检索任务上得分为 55.26。平台 API 兼容通用聊天界面，提供 Python 与 Javascript 客户端库，并支持系统提示词。服务已开放注册，容量将逐步提升。

## 正文

Mistral AI brings the strongest open generative models to the developers, along with efficient ways to deploy and customise them for production.

We're opening a beta access to our first platform services today. We start simple: la plateforme serves three chat endpoints for generating text following textual instructions and an embedding endpoint. Each endpoint has a different performance/price tradeoff.

The two first endpoints, mistral-tiny and mistral-small, currently use our two released open models; the third, mistral-medium, uses a prototype model with higher performances that we are testing in a deployed setting.

We serve instructed versions of our models. We have worked on consolidating the most effective alignment techniques (efficient fine-tuning, direct preference optimisation) to create easy-to-control and pleasant-to-use models. We pre-train models on data extracted from the open Web and perform instruction fine-tuning from annotations.

Mistral-tiny. Our most cost-effective endpoint currently serves Mistral 7B Instruct v0.2, a new minor release of Mistral 7B Instruct. Mistral-tiny only works in English. It obtains 7.6 on MT-Bench. The instructed model can be downloaded here .

Mistral-small. This endpoint currently serves our newest model, Mixtral 8x7B, described in more detail in our blog post . It masters English/French/Italian/German/Spanish and code and obtains 8.3 on MT-Bench.

Mistral-medium. Our highest-quality endpoint currently serves a prototype model, that is currently among the top serviced models available based on standard benchmarks. It masters English/French/Italian/German/Spanish and code and obtains a score of 8.6 on MT-Bench. The following table compare the performance of the base models of Mistral-medium, Mistral-small and the endpoint of a competitor.

Mistral-embed, our embedding endpoint, serves an embedding model with a 1024 embedding dimension. Our embedding model has been designed with retrieval capabilities in mind. It achieves a retrieval score of 55.26 on MTEB.

Our API follows the specifications of the popular chat interface initially proposed by our dearest competitor. We provide a Python and Javascript client library to query our endpoints. Our endpoints allow users to provide a system prompt to set a higher level of moderation on model outputs for applications where this is an important requirement.

Anyone can register to use our API as of today as we progressively ramp up our capacity. Our business team can help qualify your needs and accelerate access. Expect rough edges as we stabilise our platform towards fully self-served availability.

We are grateful to NVIDIA for supporting us in integrating TensorRT-LLM and Triton and working alongside us to make a sparse mixture of experts compatible with TRT-LLM.

0%