 Alexander Korolev
|
183e6cfea5
upstream-watch 10-17: SGLang check — v0.5.21 has no MuseSpark/Step5 either; but MuseSpark landing via PR #44530 5-layer series (#44664 Docker + #44665 kernels merged, model layer in-flight) + merged cookbook; SGLang likely runs Muse-816B before vLLM. SGLang has bailing_moe_v3 (Ling) as an alt engine option
|
il y a 6 jours |
|
Alexander Korolev
|
52fa899b5b
upstream-watch 10-17: Muse Spark Open 816B NVFP4 research — 497GB omni (text+img+video+audio), 1M ctx, NVFP4 g16; BLOCKED: MuseSpark arch not in vLLM v0.31.0/main (only MuseGlimmer 30B), PR #62064 is sm80+INT4 not NVFP4/SM121; ≥700GB VRAM; runnable Muse alt = Muse-Glimmer-30B (on NAS, vLLM-supported)
|
il y a 6 jours |
|
Alexander Korolev
|
74ece11aca
upstream-watch 2026-10-17: v0.32.0 still rc (base stays v0.31.0); NVIDIA FP4-Local family (GLM-5.3/Qwen3.8-Flash-Next/Nemotron) all need b12x container (new lossless NVFP4-scale fmt); GB10 stuck-clock bug (386060) + GPU-mem-not-cgroup-contained (386057) — both bit us; Ling fp4 converter in progress
|
il y a 6 jours |
