GoThread
appearance log in
home all local communities
P

Pumpkin Escobar

@PumpkinEscobar@lemmy.world

member since 12 Jun 2023 20:46 · modlog

postscomments
  • on Is anyone self-hosting their own AI? in c/selfhosted@lemmy.world · 17 Oct 2026

    Qwen 3.8 flash next on a 4090, using strata instead of llama.cpp, gives me about 45-55 tokens/sec. Mainly used for coding tasks. The model size is perfect. If the qwen 4.0 models can be a bit less wordy in reasoning but the same size and quality they’ll be really strong.

GoThread · self-hosted threadiverse · instances · modlog · rss · appearance · AGPL-3.0