A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 12, 12:15 UTC
[ROCm][CI] Fix test_cpu_offloading for ROCm (#29548)
Signed-off-by: Micah Williamson <micah.williamson@amd.com>
Micah Williamson committed
Nov 27, 2025 at 01:54 UTC
43c5792592d9beb02eea57730ce5a4647dc0c838
1 file changed
+2
tests/v1/kv_offload/test_cpu_offloading.py
+2
@@ -20,6 +20,8 @@ ATTN_BACKENDS = ["FLASH_ATTN"]
20
21
if current_platform.is_cuda():
22
ATTN_BACKENDS.append("FLASHINFER")
23
+elif current_platform.is_rocm():
24
+ ATTN_BACKENDS = ["TRITON_ATTN"]
25
26
27
class MockSubscriber: