Post #1165687
2026-04-15 03:48 UTC
Anyone tried to selfhost LLM inference on specialised hardware, for a claude code-like usecase?
https://www.ovhcloud.com/en/public-cloud/ai-endpoints/
Or
https://endpoints.huggingface.co/
Or else?
#LLM #OpenSource #selfhosted #buildinpublic
Replies (1)
-
@km6ecc@mastodon.radio 2026-04-16 18:30
@lexoyo the quantized open weights models that will run fantastic on a 2080 Super are becoming compelling...