huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-05-28 17:52:08 +00:00
Code Issues Packages Projects Releases Wiki Activity

chore(trtllm): ensure max throughput scheduling policy is selected

Browse Source
This commit is contained in:
Morgan Funtowicz 2024-10-21 23:40:54 +02:00
parent a6ac2741a3
commit 47d8c53dda
1 changed files with 1 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
backends/trtllm/lib/backend.cpp
View File

@ -53,6 +53,7 @@ tle::ExecutorConfig huggingface::tgi::backends::GetExecutorConfig(const json &co
// Define some configuration variables
execConfig.setKvCacheConfig(tle::KvCacheConfig(true));
execConfig.setEnableChunkedContext(computeCapabilities.isPostAmpere());
execConfig.setSchedulerConfig(tle::SchedulerConfig(tle::CapacitySchedulerPolicy::kMAX_UTILIZATION));
return execConfig;
}