diff --git a/recipes/glm-5-3-flash-exlr8-exl3-experts-on-dgx-spark-tp1-4/recipe.yaml b/recipes/glm-5-3-flash-exlr8-exl3-experts-on-dgx-spark-tp1-4/recipe.yaml
index 4c866584e..030ae618f 100644
--- a/recipes/glm-5-3-flash-exlr8-exl3-experts-on-dgx-spark-tp1-4/recipe.yaml
+++ b/recipes/glm-5-3-flash-exlr8-exl3-experts-on-dgx-spark-tp1-4/recipe.yaml
@@ -42,9 +42,9 @@ tasks:
   - code
   - vision
 storage_requirements:
-  nvme_gb: 700
-  ram_gb: 128
-  note: 'NVMe figure is for the assembled release; nodes pull overlays rather than copying the full weight set.'
+  ram_gb: 128.0
+  nvme_gb: 182.0
+  note: 'Weights 181.3 GiB for the NVFP4 checkpoint (NOTES.md:73, safetensors indexes). TP=1 holds the full expert set; plan ~182 GiB local disk.'
 layouts:
   - tp
 max_context_len: 524288