Skip to content

LLM inference in C/C++

C++ ggml

ci : use hf-jobs-cpu-performance, disable pytest workers (#29369)

Georgi Gerganov70596c4dcb0a
diff --git a/.github/workflows/server-sanitize.yml b/.github/workflows/server-sanitize.yml
index 8ff285a42a..69777b6a2b 100644
--- a/.github/workflows/server-sanitize.yml
+++ b/.github/workflows/server-sanitize.yml
@@ -45,7 +45,7 @@ concurrency:
jobs:
server:
- runs-on: hf-jobs-cpu-xl
+ runs-on: hf-jobs-cpu-performance
strategy:
matrix:
@@ -116,7 +116,7 @@ jobs:
run: |
source .venv/bin/activate
cd tools/server/tests
- PYTEST_WORKERS=1 ./tests.sh
+ PYTEST_WORKERS=4 ./tests.sh
- name: Slow tests
id: server_integration_tests_slow
@@ -124,4 +124,4 @@ jobs:
run: |
source .venv/bin/activate
cd tools/server/tests
- PYTEST_WORKERS=1 SLOW_TESTS=1 ./tests.sh
+ PYTEST_WORKERS=4 SLOW_TESTS=1 ./tests.sh