|
|
@@ -5,7 +5,9 @@
|
|
|
# a dequantized bf16 copy (exact value reconstruction of the MXFP8 tensors)
|
|
|
# packaged as local-inference-lab/GLM-5.3-Flash-DFlash2-bf16-lab (arch
|
|
|
# DFlashDraftModel) in the shared NAS hub. The aiden-image lane uses the
|
|
|
-# sibling -bf16 repo (arch DFlash2DraftModel).
|
|
|
+# sibling -bf16 repo (arch DFlash2DraftModel). Hand-rolled HF hub repos: refs/main
|
|
|
+# must be the snapshot dir name as EXACT 40-hex, no trailing newline (the hub
|
|
|
+# offline resolver reads it unstripped — `echo` breaks resolution, use printf).
|
|
|
# GLM-5.3-Flash NVFP4 — local-inference-lab quant (vLLM) — 2× DGX Spark, TP=2.
|
|
|
#
|
|
|
# local-inference-lab mixed quant: NVFP4 routed experts (layers 3–44,
|
