 Alexander Korolev
|
8e49f12e5a
vision 2x: parked note — B12X sparse o_proj DeepGEMM fp8_einsum assert at TP2 (his verified values); 4x DCP2 is the profile lane
|
hai 1 mes |
|
Alexander Korolev
|
33cde9c3ef
vision 2x: his verified TP2 values (GMU 0.85, KV pin off) — 0.80+pin hit a DeepGEMM layout assert
|
hai 1 mes |
|
Alexander Korolev
|
923e6b2bfc
vision recipes: args block was absorbed into [mods] table (TOML section scoping) — moved top-level
|
hai 1 mes |
|
Alexander Korolev
|
2b52f97ac3
aiden lanes: drop manual HF mounts (launcher auto-mounts ~/.cache/huggingface = shared NAS on all 4 nodes); GLM 4x -> NVFP4 (in shared cache, +23% vs FP8)
|
hai 1 mes |
|
Alexander Korolev
|
01c4ccad57
vision aiden lanes: flag route with --entrypoint sh override (baked-in env-script doubles the tail); 4x gets same fix + shared NAS HF mount
|
hai 1 mes |
|
Alexander Korolev
|
7497729957
vision 2x: MODEL_PATH -> cached snapshot (his script's unconditional --revision '' breaks offline resolution)
|
hai 1 mes |
|
Alexander Korolev
|
f8ab43bcd5
vision 2x: MODEL_PATH -> cached snapshot (his script's unconditional --revision '' breaks offline resolution)
|
hai 1 mes |
|
Alexander Korolev
|
8c292e8a5d
vision 2x: point SPARK_HF_HOME at the shared NAS HF cache (DFlash2/DSpark drafts live there; the NVMe copy lacked the DSpark draft)
|
hai 1 mes |
|
Alexander Korolev
|
b2ec9376cc
vision 2x: drop the revision pin (offline refs/main resolves; pinned hash broke the draft lookup)
|
hai 1 mes |
|
Alexander Korolev
|
dc13bd0afe
vision 2x: aiden image via HIS env-script entrypoint — mods derive NODE_RANK/MASTER_ADDR from VLLM_HOST_IP + sed off the $@ tail (env-driven contract, launcher-tail dropped)
|
hai 1 mes |
|
Alexander Korolev
|
f2ba69ee5d
vision 2x: use his /opt/vllm-src/.venv python (dist-packages wheel lacks the multimodal wrap) + his env block
|
hai 1 mes |
|
Alexander Korolev
|
41f2808d7a
vision 2x: cudagraph cap from sizes list (pydantic consistency); MAX env kept
|
hai 1 mes |
|
Alexander Korolev
|
a0c57b9e2f
aiden lanes: entrypoint=[] for built-in-entrypoint images (launcher doubles CMD otherwise); vision 2x uses proven overlay img + aiden DSpark-greedy additions (his env-script needs a dedicated pass)
|
hai 1 mes |
|
Alexander Korolev
|
ce4727d2a3
aiden lanes: 2x+4x recipes from aidenle.com guides (glm prod-1.1 DFlash2-7 + dsv4 vision prod-3.72 DSpark-6); 2x to verify, 4x deferred (FP8 ckpt downloading)
|
hai 1 mes |
