c/localllama · by Xylight@lemdro.id · 187dQwen 3.5 is coming https://github.com/ggml-org/llama.cpp/pull/19468Looks to be 35b MoE and 9b dense 32 points · 2 comments · view on lemmy.world
2 Comments
Sims@lemmy.ml · 3 pts · 187d
Holy crap they have momentum !
humanspiral@lemmy.ca · 1 pts · 186d
Qwen has been the only models at such sizes with competence in programming, and so big deal.
pkjqpg1h@lemmy.zip · 1 pts · 185d