# אירוח עצמי של הטמעות עבור DiscourseAI

**URL:** https://meta.discourse.org/t/self-hosting-embeddings-for-discourseai/290925
**Category:** Self-Hosting
**Tags:** ai, ai-search, related-topics
**Created:** [8 בינואר,‏ 2024,‏ 8:49pm UTC](https://meta.discourse.org/t/self-hosting-embeddings-for-discourseai/290925 "2024-01-08T20:49:12Z")
**Posts on this page:** 1
**Showing post:** 18

<div class="post-metadata">

### Author: ![michaelfeil](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/michaelfeil/32/477406_2.png) [@michaelfeil](https://meta.discourse.org/u/michaelfeil)
#### Post date: [31 בדצמבר,‏ 2024,‏ 2:45pm UTC](https://meta.discourse.org/t/self-hosting-embeddings-for-discourseai/290925/18 "2024-12-31T14:45:07Z")

</div>

A friend just DM’ed me this thread.

Some Pro/Con’s:

- infinity supports multi-modal embeddings (aka send images/audio) to the
- amd gpu support
- multiple models supported in the same container (control the model via `model` param).
- more dtypes e.g. int8 quantization of the weights (mostly this is irrelevant, activation memory is larger)
- new models often come out via “custom modeling code” shipped in the huggingface repo. Infinity reads this pytorch code if needed. This will help you avoid “can you support xyz models” on a ongoing basis)
- more models supported (e.g. debertav2 for mixedbread)

Cons:

- cold start time of TEI is better

---

_[View the full topic](https://meta.discourse.org/t/self-hosting-embeddings-for-discourseai/290925)._