# Generating 'Llava Api Keys' is a problem

**URL:** <https://meta.discourse.org/t/generating-llava-api-keys-is-a-problem/299769>\
**Category:** Product Support\
**Tags:** ai\
**Created:** [March 18, 2024, 7:36am UTC](https://meta.discourse.org/t/generating-llava-api-keys-is-a-problem/299769 "2024-03-18T07:36:16Z")\
**Posts on this page:** 1\
**Showing post:** 5

<div class="post-metadata">

**Author:** ![Falco](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/falco/32/179432_2.png) [@Falco](https://meta.discourse.org/u/Falco)\
**Post date:** [March 18, 2024, 3:39pm UTC](https://meta.discourse.org/t/generating-llava-api-keys-is-a-problem/299769/5 "2024-03-18T15:39:17Z")

</div>

For LLaVa, we only support self-hosting via the `ghcr.io/xfalcox/llava:latest` container image at the moment.

If you do have access to a server with GPU with at least 24GB VRAM, you can self-host it, otherwise I recommend sticking to GPT-4V.

---

_[View the full topic](https://meta.discourse.org/t/generating-llava-api-keys-is-a-problem/299769)._