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 16, 07:00 UTC
[ci/build] Combine nightly and optional (#10465)
Kevin H. Luu committed
Nov 19, 2024 at 19:36 UTC
ed701ca9637306a44ba8403ba9e85be024e0dafd
1 file changed
+5
-4
.buildkite/test-pipeline.yaml
+5
-4
@@ -9,8 +9,7 @@
9
# label(str): the name of the test. emoji allowed.
10
# fast_check(bool): whether to run this on each commit on fastcheck pipeline.
11
# fast_check_only(bool): run this test on fastcheck pipeline only
12
-# nightly(bool): run this test in nightly pipeline only
13
-# optional(bool): never run this test by default (i.e. need to unblock manually)
12
+# optional(bool): never run this test by default (i.e. need to unblock manually) unless it's scheduled nightly run.
13
# command(str): the single command to run for tests. incompatible with commands.
14
# commands(list): the list of commands to run for test. incompatbile with command.
15
# mirror_hardwares(list): the list of hardwares to run the test on as well. currently only supports [amd]
@@ -336,7 +335,7 @@ steps:
335
- pytest -v -s models/embedding/vision_language -m core_model
336
337
- label: Language Models Test (Extended) # 50min
339
- nightly: true
338
+ optional: true
339
source_file_dependencies:
340
- vllm/
341
- tests/models/decoder_only/language
@@ -362,7 +361,7 @@ steps:
361
- pytest -v -s models/encoder_decoder/vision_language -m core_model
362
363
- label: Multi-Modal Models Test (Extended) # 1h15m
365
- nightly: true
364
+ optional: true
365
source_file_dependencies:
366
- vllm/
367
- tests/models/decoder_only/audio_language
@@ -513,6 +512,7 @@ steps:
512
513
- label: Distributed Tests (A100) # optional
514
gpu: a100
515
+ optional: true
516
num_gpus: 4
517
source_file_dependencies:
518
- vllm/
@@ -526,6 +526,7 @@ steps:
526
527
- label: LM Eval Large Models # optional
528
gpu: a100
529
+ optional: true
530
num_gpus: 4
531
working_dir: "/vllm-workspace/.buildkite/lm-eval-harness"
532
source_file_dependencies: