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 13, 10:45 UTC
[TPU] Rename path to tpu platform (#28452)
Signed-off-by: Kyuyeun Kim <kyuyeunk@google.com>
Kyuyeun Kim committed
Nov 11, 2025 at 11:16 UTC
df4d3a44a83681feea723cc4c4ebe9085d29d58d
1 file changed
+1
-1
vllm/platforms/__init__.py
+1
-1
@@ -38,7 +38,7 @@ def tpu_platform_plugin() -> str | None:
38
# Check for Pathways TPU proxy
39
if envs.VLLM_TPU_USING_PATHWAYS:
40
logger.debug("Confirmed TPU platform is available via Pathways proxy.")
41
- return "tpu_inference.platforms.tpu_jax.TpuPlatform"
41
+ return "tpu_inference.platforms.tpu_platform.TpuPlatform"
42
43
# Check for libtpu installation
44
try: