Merge pull request #1101 from kvcache-ai/Atream-patch-5

Update llama4.md

Merge pull request #1101 from kvcache-ai/Atream-patch-5
Update llama4.md
92a67ab5 · Atream · GitHub · 9fad782c · 98dbdcd6 · 92a67ab5
Unverified Commit 92a67ab5 authored Apr 09, 2025 by Atream Committed by GitHub Apr 09, 2025
Show whitespace changes
Inline Side-by-side

Showing with 6 additions and 0 deletions

doc/en/llama4.md doc/en/llama4.md +6 -0

No files found.
--- a/doc/en/llama4.md
+++ b/doc/en/llama4.md
@@ -25,6 +25,12 @@ https://github.com/user-attachments/assets/449706f1-784b-4931-b2ba-07687c1aca54

 ------

+## Resource Requirements
+
+The Scout model running with 16 Experts requires approximately 65 GB of memory and 10 GB of GPU memory, while the Maverick model with 128 Experts requires approximately 270 GB of memory and 12 GB of GPU memory.
+
+------
+
 ## ⚙️ Usage Instructions

 ### 1. Clone `support-llama4` Branch