fix: num-gpu-blocks-override 4000→5500 — 4000 blocks can't fit a 70K prompt

5500 blocks × 16 tokens = 88K token capacity. Covers max-model-len=80000 + output generation headroom.
2026-08-13 13:25:23 +00:00
parent cf7824313f
commit 71644e1530
1 changed files with 1 additions and 1 deletions
--- a/computility-run.yaml
+++ b/computility-run.yaml
@@ -33,7 +33,7 @@ command:
    - --dtype
    - half
    - --num-gpu-blocks-override
-    - '4000'
+    - '5500'
 env:
    - name: VLLM_ENGINE_ITERATION_TIMEOUT_S
      value: 3600