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 5, 09:45 UTC
[XPU] Skip VLLM_BATCH_INVARIANT for XPU in EAGLE DP test (#39164)
Signed-off-by: sihao.li <sihao.li@intel.com> Co-authored-by: Kunshang Ji <kunshang.ji@intel.com>
sihao_li committed
Apr 9, 2026 at 12:45 UTC
e80e633927ab11570cc4d07a55e5420d05c7fe99
1 file changed
+1
-1
tests/v1/distributed/test_eagle_dp.py
+1
-1
@@ -37,7 +37,7 @@ else:
37
strict=False,
38
)
39
async def test_run_eagle_dp(monkeypatch: pytest.MonkeyPatch, attn_backend: str):
40
- if not current_platform.is_rocm():
40
+ if not current_platform.is_rocm() and not current_platform.is_xpu():
41
# This test checks that running a model with and without eagle
42
# leads to identical tokens.
43
#