[V0 deprecation] Remove _VLLM_V1 suffixes from attention backend names (#25489)

Signed-off-by: Matthew Bonanni <mbonanni@redhat.com> Signed-off-by: Matthew Bonanni <mbonanni001@gmail.com>
2025-09-25 13:37:50 -04:00
parent 71b25b0d48
commit 3468f17ebe
42 changed files with 131 additions and 174 deletions
--- a/tests/v1/cudagraph/test_cudagraph_mode.py
+++ b/tests/v1/cudagraph/test_cudagraph_mode.py
@@ -89,7 +89,7 @@ backend_configs = {
    # Triton Attention
    "TritonAttn":
    BackendConfig(name="TritonAttn",
-                  env_vars={"VLLM_ATTENTION_BACKEND": "TRITON_ATTN_VLLM_V1"},
+                  env_vars={"VLLM_ATTENTION_BACKEND": "TRITON_ATTN"},
                  comp_config={
                      "cudagraph_mode": "FULL_AND_PIECEWISE",
                  }),