[图片]Concerns over personal privacy with the AI plugin Discourse AI was built agnostic from AI providers since day one, so the admins can pick an AI provider that they trust with the data I’ve been struggling to set up Embeddings with Mistral AI, I suspect because Mistral requires a model to be passed. Do you know whether this is possible (and if so, how), or what should be done to make it possible?

Try setting mistral-embed in the “Model name” field, that appears after you select “Provider” as OpenAI.

Thanks, that works :+1: I’m struggling to find out what would be the best tokenizer to use for this use case though. The Mixtral tokenizer is not selectable here. Do you have any suggestions?

Your post above token length according to some tokenizers: OpenAI: 45 Mixtral: 52 Gemini: 47 E5: 50 bge-large-en: 49 bge-m3: 50 mpnet: 49 Looks like Mistral-embed doesn’t differ much from the others. And since it supports a very large context window of 8k, you should be safe picking any and leaving some room to spare by limiting the context window in Discourse to 7 or 7.5k.

Looks like mistral-embed uses the same tokenizer as the first Mixtral model, and we already ship that anyway, so what do you think about enabling that tokenizer in the embeddings config page @Roman ?

Sure. I don’t see why not if it’s already there. This change will add it to the available options: https://github.com/discourse/discourse-ai/pull/1258

如何使用嵌入实现Mistral

产品支持

Roman (Roman Rizzi) 2025 年4 月 11 日 14:44 6

当然。如果已经存在，我不认为有什么问题。此更改将把它添加到可用选项中：

话题		回复	浏览量
Use Mistral for embeddings Feature ai	1	191	2026 年4 月 6 日
Support for Mistral API Feature ai	1	625	2023 年12 月 26 日
Self-Hosting Embeddings for DiscourseAI Self-Hosting ai , ai-search , related-topics	21	3316	2025 年4 月 14 日
Can´t set ai embedding model Product Support ai	3	250	2025 年7 月 16 日
Discourse AI - Embeddings Site Management ai , ai-search , related-topics	22	7730	2026 年9 月 25 日

如何使用嵌入实现Mistral

相关话题