Re-enable flashinfer_cutlass

2026-04-16 16:40:56 -07:00
parent 6b7f8dace6
commit d49fac1b8b
2 changed files with 8 additions and 17 deletions
--- a/recipes/qwen3.5-397b-int4-autoround.yaml
+++ b/recipes/qwen3.5-397b-int4-autoround.yaml
@@ -55,6 +55,7 @@ command: |
    --max-num-batched-tokens {max_num_batched_tokens} \
    --trust-remote-code \
    --chat-template unsloth.jinja \
+    --load-format instanttensor \
    -tp {tensor_parallel} \
    --distributed-executor-backend ray