[AMD] Allow aiter attention backend for Gemma-4 (#38758)

2026-09-11 18:48:26 -07:00
parent e1d364dd1f
commit 0d1bea77da
2 changed files with 22 additions and 2 deletions
@@ -548,12 +548,13 @@ def handle_model_specific_adjustments(server_args: Any):
            "ascend",
            "intel_xpu",
            "intel_amx",
+            "aiter",
        )
        assert (
            prefill_backend in accepted_backends and decode_backend in accepted_backends
        ), (
-            "Gemma4 only supports trtllm_mha, triton, ascend, intel_xpu, or intel_amx "
-            f"attention backend, got prefill={prefill_backend}, decode={decode_backend}"
+            "Gemma4 only supports trtllm_mha, triton, ascend, intel_xpu, intel_amx, or "
+            f"aiter attention backend, got prefill={prefill_backend}, decode={decode_backend}"
        )

        # The quantization/moe_runner_backend resolution moved to the override