biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Hardware][Intel GPU] Add Intel GPU(XPU) inference backend (#3814)

Browse Source 
Co-authored-by: Jiang Li <jiang1.li@intel.com>
Co-authored-by: Abhilash Majumder <abhilash.majumder@intel.com>
Co-authored-by: Abhilash Majumder <30946547+abhilash1910@users.noreply.github.com>
This commit is contained in:
Kunshang Ji
2024-06-18 02:01:25 +08:00
committed by GitHub
parent 1f12122b17
commit 728c4c8a06
31 changed files with 1998 additions and 24 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
vllm/engine/async_llm_engine.py
View File

@@ -383,6 +383,17 @@ class AsyncLLMEngine:
"Distributed execution is not supported with the CPU backend.")
from vllm.executor.cpu_executor import CPUExecutorAsync
executor_class = CPUExecutorAsync
elif engine_config.device_config.device_type == "xpu":
if distributed_executor_backend is None:
from vllm.executor.xpu_executor import XPUExecutorAsync
executor_class = XPUExecutorAsync
elif distributed_executor_backend == "ray":
initialize_ray_cluster(engine_config.parallel_config)
from vllm.executor.ray_xpu_executor import RayXPUExecutorAsync
executor_class = RayXPUExecutorAsync
else:
raise RuntimeError(
"Not supported distributed execution model on XPU device.")
elif distributed_executor_backend == "ray":
initialize_ray_cluster(engine_config.parallel_config)
from vllm.executor.ray_gpu_executor import RayGPUExecutorAsync