Revert "fix: remove --enable-chunked-prefill — conflicts with small max_num_batched_tokens"

This reverts commit 15ad56a454.
2026-08-13 16:35:49 +00:00
parent 15ad56a454
commit 048302bd4a
1 changed files with 2 additions and 1 deletions
--- a/computility-run.yaml
+++ b/computility-run.yaml
@@ -19,7 +19,8 @@ command:
    - --disable-log-requests
    - --disable-frontend-multiprocessing
    - --max-num-batched-tokens
-    - '2048'
+    - '256'
+    - --enable-chunked-prefill
    - --max-seq-len-to-capture
    - '32768'
    - --enable-auto-tool-choice