Skip to content

GitLab

"...git@developer.sourcefind.cn:2222/OpenDAS/vllm_cscc.git" did not exist on "7025b11d949b4efeb2584690c35f919c77027368"
Commit 38d80967 authored by zhuwenwen's avatar zhuwenwen
Browse files

Merge tag 'v0.10.2rc2' into v0.10.2rc2-ori

parents 33650733 880c741b
Hide whitespace changes
Inline Side-by-side
Showing with 1447 additions and 194 deletions
.buildkite/check-wheel-size.py
View file @ 38d80967
...@@ -5,11 +5,11 @@ import os ...@@ -5,11 +5,11 @@ import os
import sys import sys
import zipfile import zipfile
# Read the VLLM_MAX_SIZE_MB environment variable, defaulting to 400 MiB # Read the VLLM_MAX_SIZE_MB environment variable, defaulting to 450 MiB
# Note that we have 400 MiB quota, please use it wisely. # Note that we have 800 MiB quota, please use it wisely.
# See https://github.com/pypi/support/issues/3792 . # See https://github.com/pypi/support/issues/6326 .
# Please also sync the value with the one in Dockerfile. # Please also sync the value with the one in Dockerfile.
VLLM_MAX_SIZE_MB = int(os.environ.get("VLLM_MAX_SIZE_MB", 400)) VLLM_MAX_SIZE_MB = int(os.environ.get("VLLM_MAX_SIZE_MB", 450))
def print_top_10_largest_files(zip_file): def print_top_10_largest_files(zip_file):
......
.buildkite/nightly-benchmarks/scripts/compare-json-results.py
View file @ 38d80967
...@@ -218,7 +218,7 @@ if __name__ == "__main__": ...@@ -218,7 +218,7 @@ if __name__ == "__main__":
"--xaxis", "--xaxis",
type=str, type=str,
default="# of max concurrency.", default="# of max concurrency.",
help="column name to use as X Axis in comparision graph", help="column name to use as X Axis in comparison graph",
) )
args = parser.parse_args() args = parser.parse_args()
......
.buildkite/nightly-benchmarks/tests/serving-tests-cpu-snc2.json
View file @ 38d80967
[ [
{ {
"test_name": "serving_llama8B_tp1_sharegpt", "test_name": "serving_llama8B_bf16_tp1_sharegpt",
"qps_list": ["inf"], "qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200], "max_concurrency_list": [12, 16, 24, 32, 64, 128, 200],
"server_environment_variables": { "server_environment_variables": {
...@@ -32,7 +32,7 @@ ...@@ -32,7 +32,7 @@
} }
}, },
{ {
"test_name": "serving_llama8B_tp2_sharegpt", "test_name": "serving_llama8B_bf16_tp2_sharegpt",
"qps_list": ["inf"], "qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200], "max_concurrency_list": [12, 16, 24, 32, 64, 128, 200],
"server_environment_variables": { "server_environment_variables": {
...@@ -64,7 +64,7 @@ ...@@ -64,7 +64,7 @@
} }
}, },
{ {
"test_name": "serving_llama8B_tp4_sharegpt", "test_name": "serving_llama8B_bf16_tp4_sharegpt",
"qps_list": ["inf"], "qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200], "max_concurrency_list": [12, 16, 24, 32, 64, 128, 200],
"server_environment_variables": { "server_environment_variables": {
...@@ -96,7 +96,7 @@ ...@@ -96,7 +96,7 @@
} }
}, },
{ {
"test_name": "serving_llama8B_tp1_random_128_128", "test_name": "serving_llama8B_bf16_tp1_random_128_128",
"qps_list": ["inf"], "qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200, 1000], "max_concurrency_list": [12, 16, 24, 32, 64, 128, 200, 1000],
"server_environment_variables": { "server_environment_variables": {
...@@ -131,7 +131,7 @@ ...@@ -131,7 +131,7 @@
} }
}, },
{ {
"test_name": "serving_llama8B_tp2_random_128_128", "test_name": "serving_llama8B_bf16_tp2_random_128_128",
"qps_list": ["inf"], "qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200, 1000], "max_concurrency_list": [12, 16, 24, 32, 64, 128, 200, 1000],
"server_environment_variables": { "server_environment_variables": {
...@@ -166,7 +166,7 @@ ...@@ -166,7 +166,7 @@
} }
}, },
{ {
"test_name": "serving_llama8B_tp4_random_128_128", "test_name": "serving_llama8B_bf16_tp4_random_128_128",
"qps_list": ["inf"], "qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200, 1000], "max_concurrency_list": [12, 16, 24, 32, 64, 128, 200, 1000],
"server_environment_variables": { "server_environment_variables": {
...@@ -198,5 +198,413 @@ ...@@ -198,5 +198,413 @@
"random-output-len": 128, "random-output-len": 128,
"num_prompts": 1000 "num_prompts": 1000
} }
},
{
"test_name": "serving_llama8B_int8_tp1_sharegpt",
"qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200],
"server_environment_variables": {
"VLLM_RPC_TIMEOUT": 100000,
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
"VLLM_CPU_SGL_KERNEL": 1,
"VLLM_CPU_KVCACHE_SPACE": 40
},
"server_parameters": {
"model": "RedHatAI/Meta-Llama-3.1-8B-Instruct-quantized.w8a8",
"tensor_parallel_size": 1,
"dtype": "bfloat16",
"distributed_executor_backend": "mp",
"block_size": 128,
"trust_remote_code": "",
"disable_log_stats": "",
"enforce_eager": "",
"max_num_batched_tokens": 2048,
"max_num_seqs": 256,
"load_format": "dummy"
},
"client_parameters": {
"model": "RedHatAI/Meta-Llama-3.1-8B-Instruct-quantized.w8a8",
"backend": "vllm",
"dataset_name": "sharegpt",
"dataset_path": "./ShareGPT_V3_unfiltered_cleaned_split.json",
"num_prompts": 200
}
},
{
"test_name": "serving_llama8B_int8_tp2_sharegpt",
"qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200],
"server_environment_variables": {
"VLLM_RPC_TIMEOUT": 100000,
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
"VLLM_CPU_SGL_KERNEL": 1,
"VLLM_CPU_KVCACHE_SPACE": 40
},
"server_parameters": {
"model": "RedHatAI/Meta-Llama-3.1-8B-Instruct-quantized.w8a8",
"tensor_parallel_size": 2,
"dtype": "bfloat16",
"distributed_executor_backend": "mp",
"block_size": 128,
"trust_remote_code": "",
"disable_log_stats": "",
"enforce_eager": "",
"max_num_batched_tokens": 2048,
"max_num_seqs": 256,
"load_format": "dummy"
},
"client_parameters": {
"model": "RedHatAI/Meta-Llama-3.1-8B-Instruct-quantized.w8a8",
"backend": "vllm",
"dataset_name": "sharegpt",
"dataset_path": "./ShareGPT_V3_unfiltered_cleaned_split.json",
"num_prompts": 200
}
},
{
"test_name": "serving_llama8B_int8_tp4_sharegpt",
"qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200],
"server_environment_variables": {
"VLLM_RPC_TIMEOUT": 100000,
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
"VLLM_CPU_SGL_KERNEL": 1,
"VLLM_CPU_KVCACHE_SPACE": 40
},
"server_parameters": {
"model": "RedHatAI/Meta-Llama-3.1-8B-Instruct-quantized.w8a8",
"tensor_parallel_size": 4,
"dtype": "bfloat16",
"distributed_executor_backend": "mp",
"block_size": 128,
"trust_remote_code": "",
"disable_log_stats": "",
"enforce_eager": "",
"max_num_batched_tokens": 2048,
"max_num_seqs": 256,
"load_format": "dummy"
},
"client_parameters": {
"model": "RedHatAI/Meta-Llama-3.1-8B-Instruct-quantized.w8a8",
"backend": "vllm",
"dataset_name": "sharegpt",
"dataset_path": "./ShareGPT_V3_unfiltered_cleaned_split.json",
"num_prompts": 200
}
},
{
"test_name": "serving_llama8B_int8_tp1_random_128_128",
"qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200, 1000],
"server_environment_variables": {
"VLLM_RPC_TIMEOUT": 100000,
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
"VLLM_CPU_SGL_KERNEL": 1,
"VLLM_CPU_KVCACHE_SPACE": 40
},
"server_parameters": {
"model": "RedHatAI/Meta-Llama-3.1-8B-Instruct-quantized.w8a8",
"tensor_parallel_size": 1,
"dtype": "bfloat16",
"distributed_executor_backend": "mp",
"block_size": 128,
"trust_remote_code": "",
"enable_chunked_prefill": "",
"disable_log_stats": "",
"enforce_eager": "",
"max_num_batched_tokens": 2048,
"max_num_seqs": 256,
"load_format": "dummy"
},
"client_parameters": {
"model": "RedHatAI/Meta-Llama-3.1-8B-Instruct-quantized.w8a8",
"backend": "vllm",
"dataset_name": "random",
"random-input-len": 128,
"random-output-len": 128,
"ignore-eos": "",
"num_prompts": 1000
}
},
{
"test_name": "serving_llama8B_int8_tp2_random_128_128",
"qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200, 1000],
"server_environment_variables": {
"VLLM_RPC_TIMEOUT": 100000,
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
"VLLM_CPU_SGL_KERNEL": 1,
"VLLM_CPU_KVCACHE_SPACE": 40
},
"server_parameters": {
"model": "RedHatAI/Meta-Llama-3.1-8B-Instruct-quantized.w8a8",
"tensor_parallel_size": 2,
"dtype": "bfloat16",
"distributed_executor_backend": "mp",
"block_size": 128,
"trust_remote_code": "",
"enable_chunked_prefill": "",
"disable_log_stats": "",
"enforce_eager": "",
"max_num_batched_tokens": 2048,
"max_num_seqs": 256,
"load_format": "dummy"
},
"client_parameters": {
"model": "RedHatAI/Meta-Llama-3.1-8B-Instruct-quantized.w8a8",
"backend": "vllm",
"dataset_name": "random",
"random-input-len": 128,
"random-output-len": 128,
"ignore-eos": "",
"num_prompts": 1000
}
},
{
"test_name": "serving_llama8B_int8_tp4_random_128_128",
"qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200, 1000],
"server_environment_variables": {
"VLLM_RPC_TIMEOUT": 100000,
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
"VLLM_CPU_SGL_KERNEL": 1,
"VLLM_CPU_KVCACHE_SPACE": 40
},
"server_parameters": {
"model": "RedHatAI/Meta-Llama-3.1-8B-Instruct-quantized.w8a8",
"tensor_parallel_size": 4,
"dtype": "bfloat16",
"distributed_executor_backend": "mp",
"block_size": 128,
"trust_remote_code": "",
"enable_chunked_prefill": "",
"disable_log_stats": "",
"enforce_eager": "",
"max_num_batched_tokens": 2048,
"max_num_seqs": 256,
"load_format": "dummy"
},
"client_parameters": {
"model": "RedHatAI/Meta-Llama-3.1-8B-Instruct-quantized.w8a8",
"backend": "vllm",
"dataset_name": "random",
"random-input-len": 128,
"random-output-len": 128,
"ignore-eos": "",
"num_prompts": 1000
}
},
{
"test_name": "serving_llama8B_int4_tp1_sharegpt",
"qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200],
"server_environment_variables": {
"VLLM_RPC_TIMEOUT": 100000,
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
"VLLM_CPU_SGL_KERNEL": 1,
"VLLM_CPU_KVCACHE_SPACE": 40
},
"server_parameters": {
"model": "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4",
"quantization": "awq",
"tensor_parallel_size": 1,
"dtype": "bfloat16",
"distributed_executor_backend": "mp",
"block_size": 128,
"trust_remote_code": "",
"disable_log_stats": "",
"enforce_eager": "",
"max_num_batched_tokens": 2048,
"max_num_seqs": 256,
"load_format": "dummy"
},
"client_parameters": {
"model": "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4",
"backend": "vllm",
"dataset_name": "sharegpt",
"dataset_path": "./ShareGPT_V3_unfiltered_cleaned_split.json",
"num_prompts": 200
}
},
{
"test_name": "serving_llama8B_int4_tp2_sharegpt",
"qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200],
"server_environment_variables": {
"VLLM_RPC_TIMEOUT": 100000,
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
"VLLM_CPU_SGL_KERNEL": 1,
"VLLM_CPU_KVCACHE_SPACE": 40
},
"server_parameters": {
"model": "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4",
"quantization": "awq",
"tensor_parallel_size": 2,
"dtype": "bfloat16",
"distributed_executor_backend": "mp",
"block_size": 128,
"trust_remote_code": "",
"disable_log_stats": "",
"enforce_eager": "",
"max_num_batched_tokens": 2048,
"max_num_seqs": 256,
"load_format": "dummy"
},
"client_parameters": {
"model": "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4",
"backend": "vllm",
"dataset_name": "sharegpt",
"dataset_path": "./ShareGPT_V3_unfiltered_cleaned_split.json",
"num_prompts": 200
}
},
{
"test_name": "serving_llama8B_int4_tp4_sharegpt",
"qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200],
"server_environment_variables": {
"VLLM_RPC_TIMEOUT": 100000,
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
"VLLM_CPU_SGL_KERNEL": 1,
"VLLM_CPU_KVCACHE_SPACE": 40
},
"server_parameters": {
"model": "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4",
"quantization": "awq",
"tensor_parallel_size": 4,
"dtype": "bfloat16",
"distributed_executor_backend": "mp",
"block_size": 128,
"trust_remote_code": "",
"disable_log_stats": "",
"enforce_eager": "",
"max_num_batched_tokens": 2048,
"max_num_seqs": 256,
"load_format": "dummy"
},
"client_parameters": {
"model": "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4",
"backend": "vllm",
"dataset_name": "sharegpt",
"dataset_path": "./ShareGPT_V3_unfiltered_cleaned_split.json",
"num_prompts": 200
}
},
{
"test_name": "serving_llama8B_int4_tp1_random_128_128",
"qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200, 1000],
"server_environment_variables": {
"VLLM_RPC_TIMEOUT": 100000,
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
"VLLM_CPU_SGL_KERNEL": 1,
"VLLM_CPU_KVCACHE_SPACE": 40
},
"server_parameters": {
"model": "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4",
"quantization": "awq",
"tensor_parallel_size": 1,
"dtype": "bfloat16",
"distributed_executor_backend": "mp",
"block_size": 128,
"trust_remote_code": "",
"enable_chunked_prefill": "",
"disable_log_stats": "",
"enforce_eager": "",
"max_num_batched_tokens": 2048,
"max_num_seqs": 256,
"load_format": "dummy"
},
"client_parameters": {
"model": "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4",
"backend": "vllm",
"dataset_name": "random",
"random-input-len": 128,
"random-output-len": 128,
"ignore-eos": "",
"num_prompts": 1000
}
},
{
"test_name": "serving_llama8B_int4_tp2_random_128_128",
"qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200, 1000],
"server_environment_variables": {
"VLLM_RPC_TIMEOUT": 100000,
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
"VLLM_CPU_SGL_KERNEL": 1,
"VLLM_CPU_KVCACHE_SPACE": 40
},
"server_parameters": {
"model": "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4",
"quantization": "awq",
"tensor_parallel_size": 2,
"dtype": "bfloat16",
"distributed_executor_backend": "mp",
"block_size": 128,
"trust_remote_code": "",
"enable_chunked_prefill": "",
"disable_log_stats": "",
"enforce_eager": "",
"max_num_batched_tokens": 2048,
"max_num_seqs": 256,
"load_format": "dummy"
},
"client_parameters": {
"model": "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4",
"backend": "vllm",
"dataset_name": "random",
"random-input-len": 128,
"random-output-len": 128,
"ignore-eos": "",
"num_prompts": 1000
}
},
{
"test_name": "serving_llama8B_int4_tp4_random_128_128",
"qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200, 1000],
"server_environment_variables": {
"VLLM_RPC_TIMEOUT": 100000,
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
"VLLM_CPU_SGL_KERNEL": 1,
"VLLM_CPU_KVCACHE_SPACE": 40
},
"server_parameters": {
"model": "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4",
"quantization": "awq",
"tensor_parallel_size": 4,
"dtype": "bfloat16",
"distributed_executor_backend": "mp",
"block_size": 128,
"trust_remote_code": "",
"enable_chunked_prefill": "",
"disable_log_stats": "",
"enforce_eager": "",
"max_num_batched_tokens": 2048,
"max_num_seqs": 256,
"load_format": "dummy"
},
"client_parameters": {
"model": "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4",
"backend": "vllm",
"dataset_name": "random",
"random-input-len": 128,
"random-output-len": 128,
"ignore-eos": "",
"num_prompts": 1000
}
} }
] ]
.buildkite/nightly-benchmarks/tests/serving-tests-cpu-snc3.json
View file @ 38d80967
[ [
{ {
"test_name": "serving_llama8B_pp1_sharegpt", "test_name": "serving_llama8B_bf16_pp1_sharegpt",
"qps_list": ["inf"], "qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200], "max_concurrency_list": [12, 16, 24, 32, 64, 128, 200],
"server_environment_variables": { "server_environment_variables": {
...@@ -32,7 +32,39 @@ ...@@ -32,7 +32,39 @@
} }
}, },
{ {
"test_name": "serving_llama8B_pp3_sharegpt", "test_name": "serving_llama8B_bf16_tp2_sharegpt",
"qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200],
"server_environment_variables": {
"VLLM_RPC_TIMEOUT": 100000,
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
"VLLM_CPU_SGL_KERNEL": 1,
"VLLM_CPU_KVCACHE_SPACE": 40
},
"server_parameters": {
"model": "meta-llama/Llama-3.1-8B-Instruct",
"tensor_parallel_size": 2,
"dtype": "bfloat16",
"distributed_executor_backend": "mp",
"block_size": 128,
"trust_remote_code": "",
"disable_log_stats": "",
"enforce_eager": "",
"max_num_batched_tokens": 2048,
"max_num_seqs": 256,
"load_format": "dummy"
},
"client_parameters": {
"model": "meta-llama/Llama-3.1-8B-Instruct",
"backend": "vllm",
"dataset_name": "sharegpt",
"dataset_path": "./ShareGPT_V3_unfiltered_cleaned_split.json",
"num_prompts": 200
}
},
{
"test_name": "serving_llama8B_bf16_pp3_sharegpt",
"qps_list": ["inf"], "qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200], "max_concurrency_list": [12, 16, 24, 32, 64, 128, 200],
"server_environment_variables": { "server_environment_variables": {
...@@ -64,7 +96,7 @@ ...@@ -64,7 +96,7 @@
} }
}, },
{ {
"test_name": "serving_llama8B_tp2pp3_sharegpt", "test_name": "serving_llama8B_bf16_tp2pp3_sharegpt",
"qps_list": ["inf"], "qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200], "max_concurrency_list": [12, 16, 24, 32, 64, 128, 200],
"server_environment_variables": { "server_environment_variables": {
...@@ -97,7 +129,7 @@ ...@@ -97,7 +129,7 @@
} }
}, },
{ {
"test_name": "serving_llama8B_pp1_random_128_128", "test_name": "serving_llama8B_bf16_pp1_random_128_128",
"qps_list": ["inf"], "qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200, 1000], "max_concurrency_list": [12, 16, 24, 32, 64, 128, 200, 1000],
"server_environment_variables": { "server_environment_variables": {
...@@ -132,7 +164,42 @@ ...@@ -132,7 +164,42 @@
} }
}, },
{ {
"test_name": "serving_llama8B_pp3_random_128_128", "test_name": "serving_llama8B_bf16_tp2_random_128_128",
"qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200, 1000],
"server_environment_variables": {
"VLLM_RPC_TIMEOUT": 100000,
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
"VLLM_CPU_SGL_KERNEL": 1,
"VLLM_CPU_KVCACHE_SPACE": 40
},
"server_parameters": {
"model": "meta-llama/Llama-3.1-8B-Instruct",
"tensor_parallel_size": 2,
"dtype": "bfloat16",
"distributed_executor_backend": "mp",
"block_size": 128,
"trust_remote_code": "",
"enable_chunked_prefill": "",
"disable_log_stats": "",
"enforce_eager": "",
"max_num_batched_tokens": 2048,
"max_num_seqs": 256,
"load_format": "dummy"
},
"client_parameters": {
"model": "meta-llama/Llama-3.1-8B-Instruct",
"backend": "vllm",
"dataset_name": "random",
"random-input-len": 128,
"random-output-len": 128,
"ignore-eos": "",
"num_prompts": 1000
}
},
{
"test_name": "serving_llama8B_bf16_pp3_random_128_128",
"qps_list": ["inf"], "qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200, 1000], "max_concurrency_list": [12, 16, 24, 32, 64, 128, 200, 1000],
"server_environment_variables": { "server_environment_variables": {
...@@ -167,7 +234,7 @@ ...@@ -167,7 +234,7 @@
} }
}, },
{ {
"test_name": "serving_llama8B_tp2pp3_random_128_128", "test_name": "serving_llama8B_bf16_tp2pp3_random_128_128",
"qps_list": ["inf"], "qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200, 1000], "max_concurrency_list": [12, 16, 24, 32, 64, 128, 200, 1000],
"server_environment_variables": { "server_environment_variables": {
...@@ -201,5 +268,553 @@ ...@@ -201,5 +268,553 @@
"ignore-eos": "", "ignore-eos": "",
"num_prompts": 1000 "num_prompts": 1000
} }
},
{
"test_name": "serving_llama8B_int8_pp1_sharegpt",
"qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200],
"server_environment_variables": {
"VLLM_RPC_TIMEOUT": 100000,
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
"VLLM_CPU_SGL_KERNEL": 1,
"VLLM_CPU_KVCACHE_SPACE": 40
},
"server_parameters": {
"model": "RedHatAI/Meta-Llama-3.1-8B-Instruct-quantized.w8a8",
"pipeline_parallel_size": 1,
"dtype": "bfloat16",
"distributed_executor_backend": "mp",
"block_size": 128,
"trust_remote_code": "",
"disable_log_stats": "",
"enforce_eager": "",
"max_num_batched_tokens": 2048,
"max_num_seqs": 256,
"load_format": "dummy"
},
"client_parameters": {
"model": "RedHatAI/Meta-Llama-3.1-8B-Instruct-quantized.w8a8",
"backend": "vllm",
"dataset_name": "sharegpt",
"dataset_path": "./ShareGPT_V3_unfiltered_cleaned_split.json",
"num_prompts": 200
}
},
{
"test_name": "serving_llama8B_int8_tp2_sharegpt",
"qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200],
"server_environment_variables": {
"VLLM_RPC_TIMEOUT": 100000,
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
"VLLM_CPU_SGL_KERNEL": 1,
"VLLM_CPU_KVCACHE_SPACE": 40
},
"server_parameters": {
"model": "RedHatAI/Meta-Llama-3.1-8B-Instruct-quantized.w8a8",
"tensor_parallel_size": 2,
"dtype": "bfloat16",
"distributed_executor_backend": "mp",
"block_size": 128,
"trust_remote_code": "",
"disable_log_stats": "",
"enforce_eager": "",
"max_num_batched_tokens": 2048,
"max_num_seqs": 256,
"load_format": "dummy"
},
"client_parameters": {
"model": "RedHatAI/Meta-Llama-3.1-8B-Instruct-quantized.w8a8",
"backend": "vllm",
"dataset_name": "sharegpt",
"dataset_path": "./ShareGPT_V3_unfiltered_cleaned_split.json",
"num_prompts": 200
}
},
{
"test_name": "serving_llama8B_int8_pp3_sharegpt",
"qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200],
"server_environment_variables": {
"VLLM_RPC_TIMEOUT": 100000,
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
"VLLM_CPU_SGL_KERNEL": 1,
"VLLM_CPU_KVCACHE_SPACE": 40
},
"server_parameters": {
"model": "RedHatAI/Meta-Llama-3.1-8B-Instruct-quantized.w8a8",
"pipeline_parallel_size": 3,
"dtype": "bfloat16",
"distributed_executor_backend": "mp",
"block_size": 128,
"trust_remote_code": "",
"disable_log_stats": "",
"enforce_eager": "",
"max_num_batched_tokens": 2048,
"max_num_seqs": 256,
"load_format": "dummy"
},
"client_parameters": {
"model": "RedHatAI/Meta-Llama-3.1-8B-Instruct-quantized.w8a8",
"backend": "vllm",
"dataset_name": "sharegpt",
"dataset_path": "./ShareGPT_V3_unfiltered_cleaned_split.json",
"num_prompts": 200
}
},
{
"test_name": "serving_llama8B_int8_tp2pp3_sharegpt",
"qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200],
"server_environment_variables": {
"VLLM_RPC_TIMEOUT": 100000,
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
"VLLM_CPU_SGL_KERNEL": 1,
"VLLM_CPU_KVCACHE_SPACE": 40
},
"server_parameters": {
"model": "RedHatAI/Meta-Llama-3.1-8B-Instruct-quantized.w8a8",
"tensor_parallel_size": 2,
"pipeline_parallel_size": 3,
"dtype": "bfloat16",
"distributed_executor_backend": "mp",
"block_size": 128,
"trust_remote_code": "",
"disable_log_stats": "",
"enforce_eager": "",
"max_num_batched_tokens": 2048,
"max_num_seqs": 256,
"load_format": "dummy"
},
"client_parameters": {
"model": "RedHatAI/Meta-Llama-3.1-8B-Instruct-quantized.w8a8",
"backend": "vllm",
"dataset_name": "sharegpt",
"dataset_path": "./ShareGPT_V3_unfiltered_cleaned_split.json",
"num_prompts": 200
}
},
{
"test_name": "serving_llama8B_int8_pp1_random_128_128",
"qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200, 1000],
"server_environment_variables": {
"VLLM_RPC_TIMEOUT": 100000,
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
"VLLM_CPU_SGL_KERNEL": 1,
"VLLM_CPU_KVCACHE_SPACE": 40
},
"server_parameters": {
"model": "RedHatAI/Meta-Llama-3.1-8B-Instruct-quantized.w8a8",
"pipeline_parallel_size": 1,
"dtype": "bfloat16",
"distributed_executor_backend": "mp",
"block_size": 128,
"trust_remote_code": "",
"enable_chunked_prefill": "",
"disable_log_stats": "",
"enforce_eager": "",
"max_num_batched_tokens": 2048,
"max_num_seqs": 256,
"load_format": "dummy"
},
"client_parameters": {
"model": "RedHatAI/Meta-Llama-3.1-8B-Instruct-quantized.w8a8",
"backend": "vllm",
"dataset_name": "random",
"random-input-len": 128,
"random-output-len": 128,
"ignore-eos": "",
"num_prompts": 1000
}
},
{
"test_name": "serving_llama8B_int8_tp2_random_128_128",
"qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200, 1000],
"server_environment_variables": {
"VLLM_RPC_TIMEOUT": 100000,
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
"VLLM_CPU_SGL_KERNEL": 1,
"VLLM_CPU_KVCACHE_SPACE": 40
},
"server_parameters": {
"model": "RedHatAI/Meta-Llama-3.1-8B-Instruct-quantized.w8a8",
"tensor_parallel_size": 2,
"dtype": "bfloat16",
"distributed_executor_backend": "mp",
"block_size": 128,
"trust_remote_code": "",
"enable_chunked_prefill": "",
"disable_log_stats": "",
"enforce_eager": "",
"max_num_batched_tokens": 2048,
"max_num_seqs": 256,
"load_format": "dummy"
},
"client_parameters": {
"model": "RedHatAI/Meta-Llama-3.1-8B-Instruct-quantized.w8a8",
"backend": "vllm",
"dataset_name": "random",
"random-input-len": 128,
"random-output-len": 128,
"ignore-eos": "",
"num_prompts": 1000
}
},
{
"test_name": "serving_llama8B_int8_pp3_random_128_128",
"qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200, 1000],
"server_environment_variables": {
"VLLM_RPC_TIMEOUT": 100000,
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
"VLLM_CPU_SGL_KERNEL": 1,
"VLLM_CPU_KVCACHE_SPACE": 40
},
"server_parameters": {
"model": "RedHatAI/Meta-Llama-3.1-8B-Instruct-quantized.w8a8",
"pipeline_parallel_size": 3,
"dtype": "bfloat16",
"distributed_executor_backend": "mp",
"block_size": 128,
"trust_remote_code": "",
"enable_chunked_prefill": "",
"disable_log_stats": "",
"enforce_eager": "",
"max_num_batched_tokens": 2048,
"max_num_seqs": 256,
"load_format": "dummy"
},
"client_parameters": {
"model": "RedHatAI/Meta-Llama-3.1-8B-Instruct-quantized.w8a8",
"backend": "vllm",
"dataset_name": "random",
"random-input-len": 128,
"random-output-len": 128,
"ignore-eos": "",
"num_prompts": 1000
}
},
{
"test_name": "serving_llama8B_int8_tp2pp3_random_128_128",
"qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200, 1000],
"server_environment_variables": {
"VLLM_RPC_TIMEOUT": 100000,
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
"VLLM_CPU_SGL_KERNEL": 1,
"VLLM_CPU_KVCACHE_SPACE": 40
},
"server_parameters": {
"model": "RedHatAI/Meta-Llama-3.1-8B-Instruct-quantized.w8a8",
"tensor_parallel_size": 2,
"pipeline_parallel_size": 3,
"dtype": "bfloat16",
"distributed_executor_backend": "mp",
"block_size": 128,
"trust_remote_code": "",
"enable_chunked_prefill": "",
"disable_log_stats": "",
"enforce_eager": "",
"max_num_batched_tokens": 2048,
"max_num_seqs": 256,
"load_format": "dummy"
},
"client_parameters": {
"model": "RedHatAI/Meta-Llama-3.1-8B-Instruct-quantized.w8a8",
"backend": "vllm",
"dataset_name": "random",
"random-input-len": 128,
"random-output-len": 128,
"ignore-eos": "",
"num_prompts": 1000
}
},
{
"test_name": "serving_llama8B_int4_pp1_sharegpt",
"qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200],
"server_environment_variables": {
"VLLM_RPC_TIMEOUT": 100000,
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
"VLLM_CPU_SGL_KERNEL": 1,
"VLLM_CPU_KVCACHE_SPACE": 40
},
"server_parameters": {
"model": "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4",
"quantization": "awq",
"pipeline_parallel_size": 1,
"dtype": "bfloat16",
"distributed_executor_backend": "mp",
"block_size": 128,
"trust_remote_code": "",
"disable_log_stats": "",
"enforce_eager": "",
"max_num_batched_tokens": 2048,
"max_num_seqs": 256,
"load_format": "dummy"
},
"client_parameters": {
"model": "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4",
"backend": "vllm",
"dataset_name": "sharegpt",
"dataset_path": "./ShareGPT_V3_unfiltered_cleaned_split.json",
"num_prompts": 200
}
},
{
"test_name": "serving_llama8B_int4_tp2_sharegpt",
"qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200],
"server_environment_variables": {
"VLLM_RPC_TIMEOUT": 100000,
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
"VLLM_CPU_SGL_KERNEL": 1,
"VLLM_CPU_KVCACHE_SPACE": 40
},
"server_parameters": {
"model": "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4",
"quantization": "awq",
"tensor_parallel_size": 2,
"dtype": "bfloat16",
"distributed_executor_backend": "mp",
"block_size": 128,
"trust_remote_code": "",
"disable_log_stats": "",
"enforce_eager": "",
"max_num_batched_tokens": 2048,
"max_num_seqs": 256,
"load_format": "dummy"
},
"client_parameters": {
"model": "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4",
"backend": "vllm",
"dataset_name": "sharegpt",
"dataset_path": "./ShareGPT_V3_unfiltered_cleaned_split.json",
"num_prompts": 200
}
},
{
"test_name": "serving_llama8B_int4_pp3_sharegpt",
"qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200],
"server_environment_variables": {
"VLLM_RPC_TIMEOUT": 100000,
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
"VLLM_CPU_SGL_KERNEL": 1,
"VLLM_CPU_KVCACHE_SPACE": 40
},
"server_parameters": {
"model": "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4",
"quantization": "awq",
"pipeline_parallel_size": 3,
"dtype": "bfloat16",
"distributed_executor_backend": "mp",
"block_size": 128,
"trust_remote_code": "",
"disable_log_stats": "",
"enforce_eager": "",
"max_num_batched_tokens": 2048,
"max_num_seqs": 256,
"load_format": "dummy"
},
"client_parameters": {
"model": "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4",
"backend": "vllm",
"dataset_name": "sharegpt",
"dataset_path": "./ShareGPT_V3_unfiltered_cleaned_split.json",
"num_prompts": 200
}
},
{
"test_name": "serving_llama8B_int4_tp2pp3_sharegpt",
"qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200],
"server_environment_variables": {
"VLLM_RPC_TIMEOUT": 100000,
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
"VLLM_CPU_SGL_KERNEL": 1,
"VLLM_CPU_KVCACHE_SPACE": 40
},
"server_parameters": {
"model": "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4",
"quantization": "awq",
"tensor_parallel_size": 2,
"pipeline_parallel_size": 3,
"dtype": "bfloat16",
"distributed_executor_backend": "mp",
"block_size": 128,
"trust_remote_code": "",
"disable_log_stats": "",
"enforce_eager": "",
"max_num_batched_tokens": 2048,
"max_num_seqs": 256,
"load_format": "dummy"
},
"client_parameters": {
"model": "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4",
"backend": "vllm",
"dataset_name": "sharegpt",
"dataset_path": "./ShareGPT_V3_unfiltered_cleaned_split.json",
"num_prompts": 200
}
},
{
"test_name": "serving_llama8B_int4_pp1_random_128_128",
"qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200, 1000],
"server_environment_variables": {
"VLLM_RPC_TIMEOUT": 100000,
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
"VLLM_CPU_SGL_KERNEL": 1,
"VLLM_CPU_KVCACHE_SPACE": 40
},
"server_parameters": {
"model": "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4",
"quantization": "awq",
"pipeline_parallel_size": 1,
"dtype": "bfloat16",
"distributed_executor_backend": "mp",
"block_size": 128,
"trust_remote_code": "",
"enable_chunked_prefill": "",
"disable_log_stats": "",
"enforce_eager": "",
"max_num_batched_tokens": 2048,
"max_num_seqs": 256,
"load_format": "dummy"
},
"client_parameters": {
"model": "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4",
"backend": "vllm",
"dataset_name": "random",
"random-input-len": 128,
"random-output-len": 128,
"ignore-eos": "",
"num_prompts": 1000
}
},
{
"test_name": "serving_llama8B_int4_tp2_random_128_128",
"qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200, 1000],
"server_environment_variables": {
"VLLM_RPC_TIMEOUT": 100000,
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
"VLLM_CPU_SGL_KERNEL": 1,
"VLLM_CPU_KVCACHE_SPACE": 40
},
"server_parameters": {
"model": "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4",
"quantization": "awq",
"tensor_parallel_size": 2,
"dtype": "bfloat16",
"distributed_executor_backend": "mp",
"block_size": 128,
"trust_remote_code": "",
"enable_chunked_prefill": "",
"disable_log_stats": "",
"enforce_eager": "",
"max_num_batched_tokens": 2048,
"max_num_seqs": 256,
"load_format": "dummy"
},
"client_parameters": {
"model": "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4",
"backend": "vllm",
"dataset_name": "random",
"random-input-len": 128,
"random-output-len": 128,
"ignore-eos": "",
"num_prompts": 1000
}
},
{
"test_name": "serving_llama8B_int4_pp3_random_128_128",
"qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200, 1000],
"server_environment_variables": {
"VLLM_RPC_TIMEOUT": 100000,
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
"VLLM_CPU_SGL_KERNEL": 1,
"VLLM_CPU_KVCACHE_SPACE": 40
},
"server_parameters": {
"model": "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4",
"quantization": "awq",
"pipeline_parallel_size": 3,
"dtype": "bfloat16",
"distributed_executor_backend": "mp",
"block_size": 128,
"trust_remote_code": "",
"enable_chunked_prefill": "",
"disable_log_stats": "",
"enforce_eager": "",
"max_num_batched_tokens": 2048,
"max_num_seqs": 256,
"load_format": "dummy"
},
"client_parameters": {
"model": "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4",
"backend": "vllm",
"dataset_name": "random",
"random-input-len": 128,
"random-output-len": 128,
"ignore-eos": "",
"num_prompts": 1000
}
},
{
"test_name": "serving_llama8B_int4_tp2pp3_random_128_128",
"qps_list": ["inf"],
"max_concurrency_list": [12, 16, 24, 32, 64, 128, 200, 1000],
"server_environment_variables": {
"VLLM_RPC_TIMEOUT": 100000,
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
"VLLM_CPU_SGL_KERNEL": 1,
"VLLM_CPU_KVCACHE_SPACE": 40
},
"server_parameters": {
"model": "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4",
"quantization": "awq",
"tensor_parallel_size": 2,
"pipeline_parallel_size": 3,
"dtype": "bfloat16",
"distributed_executor_backend": "mp",
"block_size": 128,
"trust_remote_code": "",
"enable_chunked_prefill": "",
"disable_log_stats": "",
"enforce_eager": "",
"max_num_batched_tokens": 2048,
"max_num_seqs": 256,
"load_format": "dummy"
},
"client_parameters": {
"model": "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4",
"backend": "vllm",
"dataset_name": "random",
"random-input-len": 128,
"random-output-len": 128,
"ignore-eos": "",
"num_prompts": 1000
}
} }
] ]
.buildkite/release-pipeline.yaml
View file @ 38d80967
steps: steps:
# aarch64 + CUDA builds # aarch64 + CUDA builds. PyTorch 2.8 aarch64 + CUDA wheel is only available on CUDA 12.9
- label: "Build arm64 wheel - CUDA 12.8" - label: "Build arm64 wheel - CUDA 12.9"
id: build-wheel-arm64-cuda-12-8 id: build-wheel-arm64-cuda-12-9
agents: agents:
queue: arm64_cpu_queue_postmerge queue: arm64_cpu_queue_postmerge
commands: commands:
# #NOTE: torch_cuda_arch_list is derived from upstream PyTorch build files here: # #NOTE: torch_cuda_arch_list is derived from upstream PyTorch build files here:
# https://github.com/pytorch/pytorch/blob/main/.ci/aarch64_linux/aarch64_ci_build.sh#L7 # https://github.com/pytorch/pytorch/blob/main/.ci/aarch64_linux/aarch64_ci_build.sh#L7
- "DOCKER_BUILDKIT=1 docker build --build-arg max_jobs=16 --build-arg USE_SCCACHE=1 --build-arg GIT_REPO_CHECK=1 --build-arg CUDA_VERSION=12.8.1 --build-arg torch_cuda_arch_list='8.7 9.0 10.0+PTX 12.0' --tag vllm-ci:build-image --target build --progress plain -f docker/Dockerfile ." - "DOCKER_BUILDKIT=1 docker build --build-arg max_jobs=16 --build-arg USE_SCCACHE=1 --build-arg GIT_REPO_CHECK=1 --build-arg CUDA_VERSION=12.9.1 --build-arg torch_cuda_arch_list='8.7 9.0 10.0+PTX 12.0' --tag vllm-ci:build-image --target build --progress plain -f docker/Dockerfile ."
- "mkdir artifacts" - "mkdir artifacts"
- "docker run --rm -v $(pwd)/artifacts:/artifacts_host vllm-ci:build-image bash -c 'cp -r dist /artifacts_host && chmod -R a+rw /artifacts_host'" - "docker run --rm -v $(pwd)/artifacts:/artifacts_host vllm-ci:build-image bash -c 'cp -r dist /artifacts_host && chmod -R a+rw /artifacts_host'"
- "bash .buildkite/scripts/upload-wheels.sh" - "bash .buildkite/scripts/upload-wheels.sh"
env: env:
DOCKER_BUILDKIT: "1" DOCKER_BUILDKIT: "1"
# x86 + CUDA builds - block: "Build CUDA 12.8 wheel"
key: block-build-cu128-wheel
- label: "Build wheel - CUDA 12.8" - label: "Build wheel - CUDA 12.8"
depends_on: block-build-cu128-wheel
id: build-wheel-cuda-12-8 id: build-wheel-cuda-12-8
agents: agents:
queue: cpu_queue_postmerge queue: cpu_queue_postmerge
...@@ -44,18 +47,14 @@ steps: ...@@ -44,18 +47,14 @@ steps:
env: env:
DOCKER_BUILDKIT: "1" DOCKER_BUILDKIT: "1"
# Note(simon): We can always build CUDA 11.8 wheel to ensure the build is working. # x86 + CUDA builds
# However, this block can be uncommented to save some compute hours. - label: "Build wheel - CUDA 12.9"
# - block: "Build CUDA 11.8 wheel" depends_on: ~
# key: block-build-cu118-wheel id: build-wheel-cuda-12-9
- label: "Build wheel - CUDA 11.8"
# depends_on: block-build-cu118-wheel
id: build-wheel-cuda-11-8
agents: agents:
queue: cpu_queue_postmerge queue: cpu_queue_postmerge
commands: commands:
- "DOCKER_BUILDKIT=1 docker build --build-arg max_jobs=16 --build-arg USE_SCCACHE=1 --build-arg GIT_REPO_CHECK=1 --build-arg CUDA_VERSION=11.8.0 --build-arg torch_cuda_arch_list='7.0 7.5 8.0 8.9 9.0+PTX' --tag vllm-ci:build-image --target build --progress plain -f docker/Dockerfile ." - "DOCKER_BUILDKIT=1 docker build --build-arg max_jobs=16 --build-arg USE_SCCACHE=1 --build-arg GIT_REPO_CHECK=1 --build-arg CUDA_VERSION=12.9.1 --build-arg torch_cuda_arch_list='7.0 7.5 8.0 8.9 9.0+PTX' --tag vllm-ci:build-image --target build --progress plain -f docker/Dockerfile ."
- "mkdir artifacts" - "mkdir artifacts"
- "docker run --rm -v $(pwd)/artifacts:/artifacts_host vllm-ci:build-image bash -c 'cp -r dist /artifacts_host && chmod -R a+rw /artifacts_host'" - "docker run --rm -v $(pwd)/artifacts:/artifacts_host vllm-ci:build-image bash -c 'cp -r dist /artifacts_host && chmod -R a+rw /artifacts_host'"
- "bash .buildkite/scripts/upload-wheels.sh" - "bash .buildkite/scripts/upload-wheels.sh"
...@@ -75,6 +74,7 @@ steps: ...@@ -75,6 +74,7 @@ steps:
- "docker tag public.ecr.aws/q9t5s3a7/vllm-release-repo:$BUILDKITE_COMMIT-$(uname -m) public.ecr.aws/q9t5s3a7/vllm-release-repo:$BUILDKITE_COMMIT" - "docker tag public.ecr.aws/q9t5s3a7/vllm-release-repo:$BUILDKITE_COMMIT-$(uname -m) public.ecr.aws/q9t5s3a7/vllm-release-repo:$BUILDKITE_COMMIT"
- "docker push public.ecr.aws/q9t5s3a7/vllm-release-repo:$BUILDKITE_COMMIT" - "docker push public.ecr.aws/q9t5s3a7/vllm-release-repo:$BUILDKITE_COMMIT"
# PyTorch 2.8 aarch64 + CUDA wheel is only available on CUDA 12.9
- label: "Build release image (arm64)" - label: "Build release image (arm64)"
depends_on: ~ depends_on: ~
id: build-release-image-arm64 id: build-release-image-arm64
...@@ -82,7 +82,7 @@ steps: ...@@ -82,7 +82,7 @@ steps:
queue: arm64_cpu_queue_postmerge queue: arm64_cpu_queue_postmerge
commands: commands:
- "aws ecr-public get-login-password --region us-east-1 | docker login --username AWS --password-stdin public.ecr.aws/q9t5s3a7" - "aws ecr-public get-login-password --region us-east-1 | docker login --username AWS --password-stdin public.ecr.aws/q9t5s3a7"
- "DOCKER_BUILDKIT=1 docker build --build-arg max_jobs=16 --build-arg USE_SCCACHE=1 --build-arg GIT_REPO_CHECK=1 --build-arg CUDA_VERSION=12.8.1 --build-arg torch_cuda_arch_list='8.7 9.0 10.0+PTX 12.0' --build-arg INSTALL_KV_CONNECTORS=true --tag public.ecr.aws/q9t5s3a7/vllm-release-repo:$BUILDKITE_COMMIT-$(uname -m) --target vllm-openai --progress plain -f docker/Dockerfile ." - "DOCKER_BUILDKIT=1 docker build --build-arg max_jobs=16 --build-arg USE_SCCACHE=1 --build-arg GIT_REPO_CHECK=1 --build-arg CUDA_VERSION=12.9.1 --build-arg torch_cuda_arch_list='8.7 9.0 10.0+PTX 12.0' --build-arg INSTALL_KV_CONNECTORS=true --tag public.ecr.aws/q9t5s3a7/vllm-release-repo:$BUILDKITE_COMMIT-$(uname -m) --target vllm-openai --progress plain -f docker/Dockerfile ."
- "docker push public.ecr.aws/q9t5s3a7/vllm-release-repo:$BUILDKITE_COMMIT-$(uname -m)" - "docker push public.ecr.aws/q9t5s3a7/vllm-release-repo:$BUILDKITE_COMMIT-$(uname -m)"
# Add job to create multi-arch manifest # Add job to create multi-arch manifest
...@@ -103,7 +103,7 @@ steps: ...@@ -103,7 +103,7 @@ steps:
- create-multi-arch-manifest - create-multi-arch-manifest
- build-wheel-cuda-12-8 - build-wheel-cuda-12-8
- build-wheel-cuda-12-6 - build-wheel-cuda-12-6
- build-wheel-cuda-11-8 - build-wheel-cuda-12-9
id: annotate-release-workflow id: annotate-release-workflow
agents: agents:
queue: cpu_queue_postmerge queue: cpu_queue_postmerge
...@@ -150,18 +150,24 @@ steps: ...@@ -150,18 +150,24 @@ steps:
env: env:
DOCKER_BUILDKIT: "1" DOCKER_BUILDKIT: "1"
- block: "Build Neuron release image" - label: "Build and publish nightly multi-arch image to DockerHub"
key: block-neuron-release-image-build depends_on:
depends_on: ~ - create-multi-arch-manifest
if: build.env("NIGHTLY") == "1"
- label: "Build and publish Neuron release image"
depends_on: block-neuron-release-image-build
agents: agents:
queue: neuron-postmerge queue: cpu_queue_postmerge
commands: commands:
- "aws ecr-public get-login-password --region us-east-1 | docker login --username AWS --password-stdin public.ecr.aws/q9t5s3a7" - "aws ecr-public get-login-password --region us-east-1 | docker login --username AWS --password-stdin public.ecr.aws/q9t5s3a7"
- "DOCKER_BUILDKIT=1 docker build --build-arg max_jobs=16 --build-arg GIT_REPO_CHECK=1 --tag public.ecr.aws/q9t5s3a7/vllm-neuron-release-repo:$(buildkite-agent meta-data get release-version) --tag public.ecr.aws/q9t5s3a7/vllm-neuron-release-repo:latest --progress plain -f docker/Dockerfile.neuron ." - "docker pull public.ecr.aws/q9t5s3a7/vllm-release-repo:$BUILDKITE_COMMIT"
- "docker push public.ecr.aws/q9t5s3a7/vllm-neuron-release-repo:latest" - "docker tag public.ecr.aws/q9t5s3a7/vllm-release-repo:$BUILDKITE_COMMIT vllm/vllm-openai:nightly"
- "docker push public.ecr.aws/q9t5s3a7/vllm-neuron-release-repo:$(buildkite-agent meta-data get release-version)" - "docker tag public.ecr.aws/q9t5s3a7/vllm-release-repo:$BUILDKITE_COMMIT vllm/vllm-openai:nightly-$BUILDKITE_COMMIT"
- "docker push vllm/vllm-openai:nightly"
- "docker push vllm/vllm-openai:nightly-$BUILDKITE_COMMIT"
# Clean up old nightly builds (keep only last 14)
- "bash .buildkite/scripts/cleanup-nightly-builds.sh"
plugins:
- docker-login#v3.0.0:
username: vllmbot
password-env: DOCKERHUB_TOKEN
env: env:
DOCKER_BUILDKIT: "1" DOCKER_BUILDKIT: "1"
.buildkite/scripts/cleanup-nightly-builds.sh 0 → 100755
View file @ 38d80967
#!/bin/bash
set -ex
# Clean up old nightly builds from DockerHub, keeping only the last 14 builds
# This script uses DockerHub API to list and delete old tags with "nightly-" prefix
# DockerHub API endpoint for vllm/vllm-openai repository
REPO_API_URL="https://hub.docker.com/v2/repositories/vllm/vllm-openai/tags"
# Get DockerHub token from environment
if [ -z "$DOCKERHUB_TOKEN" ]; then
echo "Error: DOCKERHUB_TOKEN environment variable is not set"
exit 1
fi
# Function to get all tags from DockerHub
get_all_tags() {
local page=1
local all_tags=""
while true; do
local response=$(curl -s -H "Authorization: Bearer $DOCKERHUB_TOKEN" \
"$REPO_API_URL?page=$page&page_size=100")
# Get both last_updated timestamp and tag name, separated by |
local tags=$(echo "$response" | jq -r '.results[] | select(.name | startswith("nightly-")) | "\(.last_updated)|\(.name)"')
if [ -z "$tags" ]; then
break
fi
all_tags="$all_tags$tags"$'\n'
page=$((page + 1))
done
# Sort by timestamp (newest first) and extract just the tag names
echo "$all_tags" | sort -r | cut -d'|' -f2
}
delete_tag() {
local tag_name="$1"
echo "Deleting tag: $tag_name"
local delete_url="https://hub.docker.com/v2/repositories/vllm/vllm-openai/tags/$tag_name"
local response=$(curl -s -X DELETE -H "Authorization: Bearer $DOCKERHUB_TOKEN" "$delete_url")
if echo "$response" | jq -e '.detail' > /dev/null 2>&1; then
echo "Warning: Failed to delete tag $tag_name: $(echo "$response" | jq -r '.detail')"
else
echo "Successfully deleted tag: $tag_name"
fi
}
# Get all nightly- prefixed tags, sorted by last_updated timestamp (newest first)
echo "Fetching all tags from DockerHub..."
all_tags=$(get_all_tags)
if [ -z "$all_tags" ]; then
echo "No tags found to clean up"
exit 0
fi
# Count total tags
total_tags=$(echo "$all_tags" | wc -l)
echo "Found $total_tags tags"
# Keep only the last 14 builds (including the current one)
tags_to_keep=14
tags_to_delete=$((total_tags - tags_to_keep))
if [ $tags_to_delete -le 0 ]; then
echo "No tags need to be deleted (only $total_tags tags found, keeping $tags_to_keep)"
exit 0
fi
echo "Will delete $tags_to_delete old tags, keeping the newest $tags_to_keep"
# Get tags to delete (skip the first $tags_to_keep tags)
tags_to_delete_list=$(echo "$all_tags" | tail -n +$((tags_to_keep + 1)))
if [ -z "$tags_to_delete_list" ]; then
echo "No tags to delete"
exit 0
fi
# Delete old tags
echo "Deleting old tags..."
while IFS= read -r tag; do
if [ -n "$tag" ]; then
delete_tag "$tag"
# Add a small delay to avoid rate limiting
sleep 1
fi
done <<< "$tags_to_delete_list"
echo "Cleanup completed successfully"
.buildkite/scripts/hardware_ci/run-neuron-test.sh deleted 100644 → 0
View file @ 33650733
#!/bin/bash
# This script build the Neuron docker image and run the API server inside the container.
# It serves a sanity check for compilation and basic model usage.
set -e
set -v
image_name="neuron/vllm-ci"
container_name="neuron_$(tr -dc A-Za-z0-9 < /dev/urandom | head -c 10; echo)"
HF_CACHE="$(realpath ~)/huggingface"
mkdir -p "${HF_CACHE}"
HF_MOUNT="/root/.cache/huggingface"
HF_TOKEN=$(aws secretsmanager get-secret-value --secret-id "ci/vllm-neuron/hf-token" --region us-west-2 --query 'SecretString' --output text | jq -r .VLLM_NEURON_CI_HF_TOKEN)
NEURON_COMPILE_CACHE_URL="$(realpath ~)/neuron_compile_cache"
mkdir -p "${NEURON_COMPILE_CACHE_URL}"
NEURON_COMPILE_CACHE_MOUNT="/root/.cache/neuron_compile_cache"
# Try building the docker image
aws ecr-public get-login-password --region us-east-1 | docker login --username AWS --password-stdin public.ecr.aws
# prune old image and containers to save disk space, and only once a day
# by using a timestamp file in tmp.
if [ -f /tmp/neuron-docker-build-timestamp ]; then
last_build=$(cat /tmp/neuron-docker-build-timestamp)
current_time=$(date +%s)
if [ $((current_time - last_build)) -gt 86400 ]; then
# Remove dangling images (those that are not tagged and not used by any container)
docker image prune -f
# Remove unused volumes / force the system prune for old images as well.
docker volume prune -f && docker system prune -f
echo "$current_time" > /tmp/neuron-docker-build-timestamp
fi
else
date "+%s" > /tmp/neuron-docker-build-timestamp
fi
docker build -t "${image_name}" -f docker/Dockerfile.neuron .
# Setup cleanup
remove_docker_container() {
docker image rm -f "${image_name}" || true;
}
trap remove_docker_container EXIT
# Run the image
docker run --rm -it --device=/dev/neuron0 --network bridge \
-v "${HF_CACHE}:${HF_MOUNT}" \
-e "HF_HOME=${HF_MOUNT}" \
-e "HF_TOKEN=${HF_TOKEN}" \
-v "${NEURON_COMPILE_CACHE_URL}:${NEURON_COMPILE_CACHE_MOUNT}" \
-e "NEURON_COMPILE_CACHE_URL=${NEURON_COMPILE_CACHE_MOUNT}" \
--name "${container_name}" \
${image_name} \
/bin/bash -c "
set -e; # Exit on first error
python3 /workspace/vllm/examples/offline_inference/neuron.py;
python3 -m pytest /workspace/vllm/tests/neuron/1_core/ -v --capture=tee-sys;
for f in /workspace/vllm/tests/neuron/2_core/*.py; do
echo \"Running test file: \$f\";
python3 -m pytest \$f -v --capture=tee-sys;
done
"
\ No newline at end of file
.buildkite/scripts/hardware_ci/run-xpu-test.sh
View file @ 38d80967
...@@ -30,10 +30,12 @@ docker run \ ...@@ -30,10 +30,12 @@ docker run \
bash -c ' bash -c '
set -e set -e
echo $ZE_AFFINITY_MASK echo $ZE_AFFINITY_MASK
VLLM_USE_V1=1 python3 examples/offline_inference/basic/generate.py --model facebook/opt-125m --block-size 64 --enforce-eager pip install tblib==3.1.0
VLLM_USE_V1=1 python3 examples/offline_inference/basic/generate.py --model facebook/opt-125m --block-size 64 -O3 -O.cudagraph_mode=NONE python3 examples/offline_inference/basic/generate.py --model facebook/opt-125m --block-size 64 --enforce-eager
VLLM_USE_V1=1 python3 examples/offline_inference/basic/generate.py --model facebook/opt-125m --block-size 64 --enforce-eager -tp 2 --distributed-executor-backend ray python3 examples/offline_inference/basic/generate.py --model facebook/opt-125m --block-size 64 -O3 -O.cudagraph_mode=NONE
VLLM_USE_V1=1 python3 examples/offline_inference/basic/generate.py --model facebook/opt-125m --block-size 64 --enforce-eager -tp 2 --distributed-executor-backend mp python3 examples/offline_inference/basic/generate.py --model facebook/opt-125m --block-size 64 --enforce-eager -tp 2 --distributed-executor-backend ray
python3 examples/offline_inference/basic/generate.py --model facebook/opt-125m --block-size 64 --enforce-eager -tp 2 --distributed-executor-backend mp
VLLM_ATTENTION_BACKEND=TRITON_ATTN_VLLM_V1 python3 examples/offline_inference/basic/generate.py --model facebook/opt-125m --block-size 64 --enforce-eager
cd tests cd tests
pytest -v -s v1/core pytest -v -s v1/core
pytest -v -s v1/engine pytest -v -s v1/engine
......
.buildkite/scripts/upload-wheels.sh
View file @ 38d80967
...@@ -58,14 +58,15 @@ python3 .buildkite/generate_index.py --wheel "$normal_wheel" ...@@ -58,14 +58,15 @@ python3 .buildkite/generate_index.py --wheel "$normal_wheel"
aws s3 cp "$wheel" "s3://vllm-wheels/$BUILDKITE_COMMIT/" aws s3 cp "$wheel" "s3://vllm-wheels/$BUILDKITE_COMMIT/"
aws s3 cp "$normal_wheel" "s3://vllm-wheels/$BUILDKITE_COMMIT/" aws s3 cp "$normal_wheel" "s3://vllm-wheels/$BUILDKITE_COMMIT/"
if [[ $normal_wheel == *"cu118"* ]]; then if [[ $normal_wheel == *"cu126"* ]]; then
# if $normal_wheel matches cu118, do not upload the index.html
echo "Skipping index files for cu118 wheels"
elif [[ $normal_wheel == *"cu126"* ]]; then
# if $normal_wheel matches cu126, do not upload the index.html # if $normal_wheel matches cu126, do not upload the index.html
echo "Skipping index files for cu126 wheels" echo "Skipping index files for cu126 wheels"
elif [[ $normal_wheel == *"cu128"* ]]; then
# if $normal_wheel matches cu128, do not upload the index.html
echo "Skipping index files for cu128 wheels"
else else
# only upload index.html for cu128 wheels (default wheels) # only upload index.html for cu129 wheels (default wheels) as it
# is available on both x86 and arm64
aws s3 cp index.html "s3://vllm-wheels/$BUILDKITE_COMMIT/vllm/index.html" aws s3 cp index.html "s3://vllm-wheels/$BUILDKITE_COMMIT/vllm/index.html"
aws s3 cp "s3://vllm-wheels/nightly/index.html" "s3://vllm-wheels/$BUILDKITE_COMMIT/index.html" aws s3 cp "s3://vllm-wheels/nightly/index.html" "s3://vllm-wheels/$BUILDKITE_COMMIT/index.html"
fi fi
...@@ -74,14 +75,15 @@ fi ...@@ -74,14 +75,15 @@ fi
aws s3 cp "$wheel" "s3://vllm-wheels/nightly/" aws s3 cp "$wheel" "s3://vllm-wheels/nightly/"
aws s3 cp "$normal_wheel" "s3://vllm-wheels/nightly/" aws s3 cp "$normal_wheel" "s3://vllm-wheels/nightly/"
if [[ $normal_wheel == *"cu118"* ]]; then if [[ $normal_wheel == *"cu126"* ]]; then
# if $normal_wheel matches cu118, do not upload the index.html
echo "Skipping index files for cu118 wheels"
elif [[ $normal_wheel == *"cu126"* ]]; then
# if $normal_wheel matches cu126, do not upload the index.html # if $normal_wheel matches cu126, do not upload the index.html
echo "Skipping index files for cu126 wheels" echo "Skipping index files for cu126 wheels"
elif [[ $normal_wheel == *"cu128"* ]]; then
# if $normal_wheel matches cu128, do not upload the index.html
echo "Skipping index files for cu128 wheels"
else else
# only upload index.html for cu128 wheels (default wheels) # only upload index.html for cu129 wheels (default wheels) as it
# is available on both x86 and arm64
aws s3 cp index.html "s3://vllm-wheels/nightly/vllm/index.html" aws s3 cp index.html "s3://vllm-wheels/nightly/vllm/index.html"
fi fi
......
.buildkite/test-pipeline.yaml
View file @ 38d80967
...@@ -41,7 +41,8 @@ steps: ...@@ -41,7 +41,8 @@ steps:
commands: commands:
- bash standalone_tests/pytorch_nightly_dependency.sh - bash standalone_tests/pytorch_nightly_dependency.sh
- label: Async Engine, Inputs, Utils, Worker Test # 24min - label: Async Engine, Inputs, Utils, Worker Test # 36min
timeout_in_minutes: 50
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
source_file_dependencies: source_file_dependencies:
- vllm/ - vllm/
...@@ -53,6 +54,7 @@ steps: ...@@ -53,6 +54,7 @@ steps:
- tests/utils_ - tests/utils_
- tests/worker - tests/worker
- tests/standalone_tests/lazy_imports.py - tests/standalone_tests/lazy_imports.py
- tests/transformers_utils
commands: commands:
- python3 standalone_tests/lazy_imports.py - python3 standalone_tests/lazy_imports.py
- pytest -v -s mq_llm_engine # MQLLMEngine - pytest -v -s mq_llm_engine # MQLLMEngine
...@@ -62,8 +64,10 @@ steps: ...@@ -62,8 +64,10 @@ steps:
- pytest -v -s multimodal - pytest -v -s multimodal
- pytest -v -s utils_ # Utils - pytest -v -s utils_ # Utils
- pytest -v -s worker # Worker - pytest -v -s worker # Worker
- pytest -v -s transformers_utils # transformers_utils
- label: Python-only Installation Test - label: Python-only Installation Test # 10min
timeout_in_minutes: 20
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
source_file_dependencies: source_file_dependencies:
- tests/standalone_tests/python_only_compile.sh - tests/standalone_tests/python_only_compile.sh
...@@ -71,7 +75,8 @@ steps: ...@@ -71,7 +75,8 @@ steps:
commands: commands:
- bash standalone_tests/python_only_compile.sh - bash standalone_tests/python_only_compile.sh
- label: Basic Correctness Test # 30min - label: Basic Correctness Test # 20min
timeout_in_minutes: 30
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
fast_check: true fast_check: true
torch_nightly: true torch_nightly: true
...@@ -88,7 +93,8 @@ steps: ...@@ -88,7 +93,8 @@ steps:
- pytest -v -s basic_correctness/test_cpu_offload.py - pytest -v -s basic_correctness/test_cpu_offload.py
- VLLM_TEST_ENABLE_ARTIFICIAL_PREEMPT=1 pytest -v -s basic_correctness/test_preemption.py - VLLM_TEST_ENABLE_ARTIFICIAL_PREEMPT=1 pytest -v -s basic_correctness/test_preemption.py
- label: Core Test # 10min - label: Core Test # 22min
timeout_in_minutes: 35
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
fast_check: true fast_check: true
source_file_dependencies: source_file_dependencies:
...@@ -98,7 +104,19 @@ steps: ...@@ -98,7 +104,19 @@ steps:
commands: commands:
- pytest -v -s core - pytest -v -s core
- label: Entrypoints Test (LLM) # 40min - label: Entrypoints Unit Tests # 5min
timeout_in_minutes: 10
working_dir: "/vllm-workspace/tests"
fast_check: true
source_file_dependencies:
- vllm/entrypoints
- tests/entrypoints/
commands:
- pytest -v -s entrypoints/openai/tool_parsers
- pytest -v -s entrypoints/ --ignore=entrypoints/llm --ignore=entrypoints/openai --ignore=entrypoints/offline_mode --ignore=entrypoints/test_chat_utils.py --ignore=entrypoints/pooling
- label: Entrypoints Integration Test (LLM) # 30min
timeout_in_minutes: 40
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
working_dir: "/vllm-workspace/tests" working_dir: "/vllm-workspace/tests"
fast_check: true fast_check: true
...@@ -114,7 +132,8 @@ steps: ...@@ -114,7 +132,8 @@ steps:
- pytest -v -s entrypoints/llm/test_generate.py # it needs a clean process - pytest -v -s entrypoints/llm/test_generate.py # it needs a clean process
- VLLM_USE_V1=0 pytest -v -s entrypoints/offline_mode # Needs to avoid interference with other tests - VLLM_USE_V1=0 pytest -v -s entrypoints/offline_mode # Needs to avoid interference with other tests
- label: Entrypoints Test (API Server) # 40min - label: Entrypoints Integration Test (API Server) # 100min
timeout_in_minutes: 130
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
working_dir: "/vllm-workspace/tests" working_dir: "/vllm-workspace/tests"
fast_check: true fast_check: true
...@@ -126,10 +145,24 @@ steps: ...@@ -126,10 +145,24 @@ steps:
commands: commands:
- export VLLM_WORKER_MULTIPROC_METHOD=spawn - export VLLM_WORKER_MULTIPROC_METHOD=spawn
- PYTHONPATH=/vllm-workspace pytest -v -s entrypoints/openai/test_collective_rpc.py # PYTHONPATH is needed to import custom Worker extension - PYTHONPATH=/vllm-workspace pytest -v -s entrypoints/openai/test_collective_rpc.py # PYTHONPATH is needed to import custom Worker extension
- pytest -v -s entrypoints/openai --ignore=entrypoints/openai/test_chat_with_tool_reasoning.py --ignore=entrypoints/openai/test_oot_registration.py --ignore=entrypoints/openai/test_tensorizer_entrypoint.py --ignore=entrypoints/openai/correctness/ --ignore=entrypoints/openai/test_collective_rpc.py - pytest -v -s entrypoints/openai --ignore=entrypoints/openai/test_chat_with_tool_reasoning.py --ignore=entrypoints/openai/test_oot_registration.py --ignore=entrypoints/openai/test_tensorizer_entrypoint.py --ignore=entrypoints/openai/correctness/ --ignore=entrypoints/openai/test_collective_rpc.py --ignore=entrypoints/openai/tool_parsers/
- pytest -v -s entrypoints/test_chat_utils.py - pytest -v -s entrypoints/test_chat_utils.py
- label: Distributed Tests (4 GPUs) # 10min - label: Entrypoints Integration Test (Pooling)
timeout_in_minutes: 50
mirror_hardwares: [amdexperimental]
working_dir: "/vllm-workspace/tests"
fast_check: true
torch_nightly: true
source_file_dependencies:
- vllm/
- tests/entrypoints/pooling
commands:
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
- pytest -v -s entrypoints/pooling
- label: Distributed Tests (4 GPUs) # 35min
timeout_in_minutes: 50
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
working_dir: "/vllm-workspace/tests" working_dir: "/vllm-workspace/tests"
num_gpus: 4 num_gpus: 4
...@@ -172,7 +205,8 @@ steps: ...@@ -172,7 +205,8 @@ steps:
- VLLM_ALLOW_INSECURE_SERIALIZATION=1 RAY_DEDUP_LOGS=0 python3 rlhf_colocate.py - VLLM_ALLOW_INSECURE_SERIALIZATION=1 RAY_DEDUP_LOGS=0 python3 rlhf_colocate.py
- popd - popd
- label: EPLB Algorithm Test - label: EPLB Algorithm Test # 5min
timeout_in_minutes: 15
working_dir: "/vllm-workspace/tests" working_dir: "/vllm-workspace/tests"
source_file_dependencies: source_file_dependencies:
- vllm/distributed/eplb - vllm/distributed/eplb
...@@ -181,6 +215,7 @@ steps: ...@@ -181,6 +215,7 @@ steps:
- pytest -v -s distributed/test_eplb_algo.py - pytest -v -s distributed/test_eplb_algo.py
- label: EPLB Execution Test # 5min - label: EPLB Execution Test # 5min
timeout_in_minutes: 15
working_dir: "/vllm-workspace/tests" working_dir: "/vllm-workspace/tests"
num_gpus: 4 num_gpus: 4
source_file_dependencies: source_file_dependencies:
...@@ -189,13 +224,14 @@ steps: ...@@ -189,13 +224,14 @@ steps:
commands: commands:
- pytest -v -s distributed/test_eplb_execute.py - pytest -v -s distributed/test_eplb_execute.py
- label: Metrics, Tracing Test # 10min - label: Metrics, Tracing Test # 12min
timeout_in_minutes: 20
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
num_gpus: 2 num_gpus: 2
source_file_dependencies: source_file_dependencies:
- vllm/ - vllm/
- tests/metrics - tests/metrics
- tests/tracing - tests/v1/tracing
commands: commands:
- pytest -v -s metrics - pytest -v -s metrics
- "pip install \ - "pip install \
...@@ -208,7 +244,8 @@ steps: ...@@ -208,7 +244,8 @@ steps:
##### fast check tests ##### ##### fast check tests #####
##### 1 GPU test ##### ##### 1 GPU test #####
- label: Regression Test # 5min - label: Regression Test # 7min
timeout_in_minutes: 20
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
source_file_dependencies: source_file_dependencies:
- vllm/ - vllm/
...@@ -218,7 +255,8 @@ steps: ...@@ -218,7 +255,8 @@ steps:
- pytest -v -s test_regression.py - pytest -v -s test_regression.py
working_dir: "/vllm-workspace/tests" # optional working_dir: "/vllm-workspace/tests" # optional
- label: Engine Test # 10min - label: Engine Test # 25min
timeout_in_minutes: 40
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
source_file_dependencies: source_file_dependencies:
- vllm/ - vllm/
...@@ -233,7 +271,8 @@ steps: ...@@ -233,7 +271,8 @@ steps:
# OOM in the CI unless we run this separately # OOM in the CI unless we run this separately
- pytest -v -s tokenization - pytest -v -s tokenization
- label: V1 Test e2e + engine - label: V1 Test e2e + engine # 30min
timeout_in_minutes: 45
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
source_file_dependencies: source_file_dependencies:
- vllm/ - vllm/
...@@ -244,7 +283,8 @@ steps: ...@@ -244,7 +283,8 @@ steps:
- pytest -v -s v1/e2e - pytest -v -s v1/e2e
- pytest -v -s v1/engine - pytest -v -s v1/engine
- label: V1 Test entrypoints - label: V1 Test entrypoints # 35min
timeout_in_minutes: 50
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
source_file_dependencies: source_file_dependencies:
- vllm/ - vllm/
...@@ -252,7 +292,8 @@ steps: ...@@ -252,7 +292,8 @@ steps:
commands: commands:
- pytest -v -s v1/entrypoints - pytest -v -s v1/entrypoints
- label: V1 Test others - label: V1 Test others # 42min
timeout_in_minutes: 60
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
source_file_dependencies: source_file_dependencies:
- vllm/ - vllm/
...@@ -276,7 +317,8 @@ steps: ...@@ -276,7 +317,8 @@ steps:
- pip install -U git+https://github.com/robertgshaw2-redhat/lm-evaluation-harness.git@streaming-api - pip install -U git+https://github.com/robertgshaw2-redhat/lm-evaluation-harness.git@streaming-api
- pytest -v -s entrypoints/openai/correctness/test_lmeval.py::test_lm_eval_accuracy_v1_engine - pytest -v -s entrypoints/openai/correctness/test_lmeval.py::test_lm_eval_accuracy_v1_engine
- label: Examples Test # 25min - label: Examples Test # 30min
timeout_in_minutes: 45
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
working_dir: "/vllm-workspace/examples" working_dir: "/vllm-workspace/examples"
source_file_dependencies: source_file_dependencies:
...@@ -294,14 +336,14 @@ steps: ...@@ -294,14 +336,14 @@ steps:
- python3 offline_inference/vision_language_pooling.py --seed 0 - python3 offline_inference/vision_language_pooling.py --seed 0
- python3 offline_inference/vision_language_multi_image.py --seed 0 - python3 offline_inference/vision_language_multi_image.py --seed 0
- VLLM_USE_V1=0 python3 others/tensorize_vllm_model.py --model facebook/opt-125m serialize --serialized-directory /tmp/ --suffix v1 && python3 others/tensorize_vllm_model.py --model facebook/opt-125m deserialize --path-to-tensors /tmp/vllm/facebook/opt-125m/v1/model.tensors - VLLM_USE_V1=0 python3 others/tensorize_vllm_model.py --model facebook/opt-125m serialize --serialized-directory /tmp/ --suffix v1 && python3 others/tensorize_vllm_model.py --model facebook/opt-125m deserialize --path-to-tensors /tmp/vllm/facebook/opt-125m/v1/model.tensors
- python3 offline_inference/encoder_decoder.py
- python3 offline_inference/encoder_decoder_multimodal.py --model-type whisper --seed 0 - python3 offline_inference/encoder_decoder_multimodal.py --model-type whisper --seed 0
- python3 offline_inference/basic/classify.py - python3 offline_inference/basic/classify.py
- python3 offline_inference/basic/embed.py - python3 offline_inference/basic/embed.py
- python3 offline_inference/basic/score.py - python3 offline_inference/basic/score.py
- VLLM_USE_V1=0 python3 offline_inference/profiling.py --model facebook/opt-125m run_num_steps --num-steps 2 - VLLM_USE_V1=0 python3 offline_inference/profiling.py --model facebook/opt-125m run_num_steps --num-steps 2
- label: Platform Tests (CUDA) - label: Platform Tests (CUDA) # 4min
timeout_in_minutes: 15
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
source_file_dependencies: source_file_dependencies:
- vllm/ - vllm/
...@@ -309,7 +351,8 @@ steps: ...@@ -309,7 +351,8 @@ steps:
commands: commands:
- pytest -v -s cuda/test_cuda_context.py - pytest -v -s cuda/test_cuda_context.py
- label: Samplers Test # 36min - label: Samplers Test # 56min
timeout_in_minutes: 75
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
source_file_dependencies: source_file_dependencies:
- vllm/model_executor/layers - vllm/model_executor/layers
...@@ -320,15 +363,23 @@ steps: ...@@ -320,15 +363,23 @@ steps:
- pytest -v -s samplers - pytest -v -s samplers
- VLLM_USE_FLASHINFER_SAMPLER=1 pytest -v -s samplers - VLLM_USE_FLASHINFER_SAMPLER=1 pytest -v -s samplers
- label: LoRA Test %N # 15min each - label: LoRA Test %N # 20min each
timeout_in_minutes: 30
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
source_file_dependencies: source_file_dependencies:
- vllm/lora - vllm/lora
- tests/lora - tests/lora
command: pytest -v -s lora --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT --ignore=lora/test_chatglm3_tp.py --ignore=lora/test_llama_tp.py --ignore=lora/test_llm_with_multi_loras.py commands:
- pytest -v -s lora \
--shard-id=$$BUILDKITE_PARALLEL_JOB \
--num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT \
--ignore=lora/test_chatglm3_tp.py \
--ignore=lora/test_llama_tp.py \
--ignore=lora/test_llm_with_multi_loras.py
parallelism: 4 parallelism: 4
- label: PyTorch Compilation Unit Tests - label: PyTorch Compilation Unit Tests # 15min
timeout_in_minutes: 30
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
torch_nightly: true torch_nightly: true
source_file_dependencies: source_file_dependencies:
...@@ -344,7 +395,8 @@ steps: ...@@ -344,7 +395,8 @@ steps:
- pytest -v -s compile/test_fusion_all_reduce.py - pytest -v -s compile/test_fusion_all_reduce.py
- pytest -v -s compile/test_decorator.py - pytest -v -s compile/test_decorator.py
- label: PyTorch Fullgraph Smoke Test # 9min - label: PyTorch Fullgraph Smoke Test # 15min
timeout_in_minutes: 30
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
torch_nightly: true torch_nightly: true
source_file_dependencies: source_file_dependencies:
...@@ -352,13 +404,10 @@ steps: ...@@ -352,13 +404,10 @@ steps:
- tests/compile - tests/compile
commands: commands:
- pytest -v -s compile/test_basic_correctness.py - pytest -v -s compile/test_basic_correctness.py
# these tests need to be separated, cannot combine - pytest -v -s compile/piecewise/
- pytest -v -s compile/piecewise/test_simple.py
- pytest -v -s compile/piecewise/test_toy_llama.py
- pytest -v -s compile/piecewise/test_full_cudagraph.py
- pytest -v -s compile/piecewise/test_multiple_graphs.py
- label: PyTorch Fullgraph Test # 18min - label: PyTorch Fullgraph Test # 20min
timeout_in_minutes: 30
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
torch_nightly: true torch_nightly: true
source_file_dependencies: source_file_dependencies:
...@@ -367,7 +416,8 @@ steps: ...@@ -367,7 +416,8 @@ steps:
commands: commands:
- pytest -v -s compile/test_full_graph.py - pytest -v -s compile/test_full_graph.py
- label: Kernels Core Operation Test - label: Kernels Core Operation Test # 48min
timeout_in_minutes: 75
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
source_file_dependencies: source_file_dependencies:
- csrc/ - csrc/
...@@ -375,7 +425,8 @@ steps: ...@@ -375,7 +425,8 @@ steps:
commands: commands:
- pytest -v -s kernels/core - pytest -v -s kernels/core
- label: Kernels Attention Test %N - label: Kernels Attention Test %N # 23min
timeout_in_minutes: 35
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
source_file_dependencies: source_file_dependencies:
- csrc/attention/ - csrc/attention/
...@@ -386,7 +437,8 @@ steps: ...@@ -386,7 +437,8 @@ steps:
- pytest -v -s kernels/attention --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT - pytest -v -s kernels/attention --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT
parallelism: 2 parallelism: 2
- label: Kernels Quantization Test %N - label: Kernels Quantization Test %N # 64min
timeout_in_minutes: 90
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
source_file_dependencies: source_file_dependencies:
- csrc/quantization/ - csrc/quantization/
...@@ -396,7 +448,8 @@ steps: ...@@ -396,7 +448,8 @@ steps:
- pytest -v -s kernels/quantization --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT - pytest -v -s kernels/quantization --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT
parallelism: 2 parallelism: 2
- label: Kernels MoE Test %N - label: Kernels MoE Test %N # 40min
timeout_in_minutes: 60
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
source_file_dependencies: source_file_dependencies:
- csrc/quantization/cutlass_w8a8/moe/ - csrc/quantization/cutlass_w8a8/moe/
...@@ -408,7 +461,8 @@ steps: ...@@ -408,7 +461,8 @@ steps:
- pytest -v -s kernels/moe --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT - pytest -v -s kernels/moe --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT
parallelism: 2 parallelism: 2
- label: Kernels Mamba Test - label: Kernels Mamba Test # 31min
timeout_in_minutes: 45
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
source_file_dependencies: source_file_dependencies:
- csrc/mamba/ - csrc/mamba/
...@@ -416,7 +470,8 @@ steps: ...@@ -416,7 +470,8 @@ steps:
commands: commands:
- pytest -v -s kernels/mamba - pytest -v -s kernels/mamba
- label: Tensorizer Test # 11min - label: Tensorizer Test # 14min
timeout_in_minutes: 25
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
source_file_dependencies: source_file_dependencies:
- vllm/model_executor/model_loader - vllm/model_executor/model_loader
...@@ -428,7 +483,8 @@ steps: ...@@ -428,7 +483,8 @@ steps:
- pytest -v -s tensorizer_loader - pytest -v -s tensorizer_loader
- pytest -v -s entrypoints/openai/test_tensorizer_entrypoint.py - pytest -v -s entrypoints/openai/test_tensorizer_entrypoint.py
- label: Model Executor Test - label: Model Executor Test # 7min
timeout_in_minutes: 20
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
source_file_dependencies: source_file_dependencies:
- vllm/model_executor - vllm/model_executor
...@@ -438,7 +494,8 @@ steps: ...@@ -438,7 +494,8 @@ steps:
- export VLLM_WORKER_MULTIPROC_METHOD=spawn - export VLLM_WORKER_MULTIPROC_METHOD=spawn
- pytest -v -s model_executor - pytest -v -s model_executor
- label: Benchmarks # 9min - label: Benchmarks # 11min
timeout_in_minutes: 20
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
working_dir: "/vllm-workspace/.buildkite" working_dir: "/vllm-workspace/.buildkite"
source_file_dependencies: source_file_dependencies:
...@@ -446,7 +503,8 @@ steps: ...@@ -446,7 +503,8 @@ steps:
commands: commands:
- bash scripts/run-benchmarks.sh - bash scripts/run-benchmarks.sh
- label: Benchmarks CLI Test # 10min - label: Benchmarks CLI Test # 7min
timeout_in_minutes: 20
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
source_file_dependencies: source_file_dependencies:
- vllm/ - vllm/
...@@ -454,7 +512,8 @@ steps: ...@@ -454,7 +512,8 @@ steps:
commands: commands:
- pytest -v -s benchmarks/ - pytest -v -s benchmarks/
- label: Quantization Test - label: Quantization Test # 70min
timeout_in_minutes: 90
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
source_file_dependencies: source_file_dependencies:
- csrc/ - csrc/
...@@ -463,10 +522,15 @@ steps: ...@@ -463,10 +522,15 @@ steps:
commands: commands:
# temporary install here since we need nightly, will move to requirements/test.in # temporary install here since we need nightly, will move to requirements/test.in
# after torchao 0.12 release, and pin a working version of torchao nightly here # after torchao 0.12 release, and pin a working version of torchao nightly here
# since torchao nightly is only compatible with torch nightly currently
# https://github.com/pytorch/ao/issues/2919, we'll have to skip new torchao tests for now
# we can only upgrade after this is resolved
- pip install --pre torchao==0.13.0.dev20250814 --index-url https://download.pytorch.org/whl/nightly/cu128 - pip install --pre torchao==0.13.0.dev20250814 --index-url https://download.pytorch.org/whl/nightly/cu128
- VLLM_TEST_FORCE_LOAD_FORMAT=auto pytest -v -s quantization - VLLM_TEST_FORCE_LOAD_FORMAT=auto pytest -v -s quantization
- label: LM Eval Small Models # 53min - label: LM Eval Small Models # 53min
timeout_in_minutes: 75
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
source_file_dependencies: source_file_dependencies:
- csrc/ - csrc/
...@@ -474,7 +538,8 @@ steps: ...@@ -474,7 +538,8 @@ steps:
commands: commands:
- pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=configs/models-small.txt --tp-size=1 - pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=configs/models-small.txt --tp-size=1
- label: OpenAI API correctness - label: OpenAI API correctness # 22min
timeout_in_minutes: 30
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
source_file_dependencies: source_file_dependencies:
- csrc/ - csrc/
...@@ -483,7 +548,8 @@ steps: ...@@ -483,7 +548,8 @@ steps:
commands: # LMEval+Transcription WER check commands: # LMEval+Transcription WER check
- pytest -s entrypoints/openai/correctness/ - pytest -s entrypoints/openai/correctness/
- label: Encoder Decoder tests # 5min - label: Encoder Decoder tests # 12min
timeout_in_minutes: 20
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
source_file_dependencies: source_file_dependencies:
- vllm/ - vllm/
...@@ -491,7 +557,8 @@ steps: ...@@ -491,7 +557,8 @@ steps:
commands: commands:
- pytest -v -s encoder_decoder - pytest -v -s encoder_decoder
- label: OpenAI-Compatible Tool Use # 20 min - label: OpenAI-Compatible Tool Use # 23 min
timeout_in_minutes: 35
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
fast_check: false fast_check: false
source_file_dependencies: source_file_dependencies:
...@@ -504,7 +571,8 @@ steps: ...@@ -504,7 +571,8 @@ steps:
##### models test ##### ##### models test #####
- label: Basic Models Test # 24min - label: Basic Models Test # 57min
timeout_in_minutes: 75
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
torch_nightly: true torch_nightly: true
source_file_dependencies: source_file_dependencies:
...@@ -517,7 +585,8 @@ steps: ...@@ -517,7 +585,8 @@ steps:
- pytest -v -s models/test_vision.py - pytest -v -s models/test_vision.py
- pytest -v -s models/test_initialization.py - pytest -v -s models/test_initialization.py
- label: Language Models Test (Standard) - label: Language Models Test (Standard) # 35min
timeout_in_minutes: 45
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
torch_nightly: true torch_nightly: true
source_file_dependencies: source_file_dependencies:
...@@ -528,6 +597,7 @@ steps: ...@@ -528,6 +597,7 @@ steps:
- pytest -v -s models/language -m core_model - pytest -v -s models/language -m core_model
- label: Language Models Test (Hybrid) # 35 min - label: Language Models Test (Hybrid) # 35 min
timeout_in_minutes: 45
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
torch_nightly: true torch_nightly: true
source_file_dependencies: source_file_dependencies:
...@@ -540,7 +610,8 @@ steps: ...@@ -540,7 +610,8 @@ steps:
- uv pip install --system --no-build-isolation 'git+https://github.com/Dao-AILab/causal-conv1d@v1.5.2' - uv pip install --system --no-build-isolation 'git+https://github.com/Dao-AILab/causal-conv1d@v1.5.2'
- pytest -v -s models/language/generation -m hybrid_model - pytest -v -s models/language/generation -m hybrid_model
- label: Language Models Test (Extended Generation) # 1hr20min - label: Language Models Test (Extended Generation) # 80min
timeout_in_minutes: 110
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
optional: true optional: true
source_file_dependencies: source_file_dependencies:
...@@ -551,7 +622,18 @@ steps: ...@@ -551,7 +622,18 @@ steps:
- pip install 'git+https://github.com/Dao-AILab/causal-conv1d@v1.5.0.post8' - pip install 'git+https://github.com/Dao-AILab/causal-conv1d@v1.5.0.post8'
- pytest -v -s models/language/generation -m '(not core_model) and (not hybrid_model)' - pytest -v -s models/language/generation -m '(not core_model) and (not hybrid_model)'
- label: Language Models Test (PPL)
timeout_in_minutes: 110
mirror_hardwares: [amdexperimental]
optional: true
source_file_dependencies:
- vllm/
- tests/models/language/generation_ppl_test
commands:
- pytest -v -s models/language/generation_ppl_test
- label: Language Models Test (Extended Pooling) # 36min - label: Language Models Test (Extended Pooling) # 36min
timeout_in_minutes: 50
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
optional: true optional: true
source_file_dependencies: source_file_dependencies:
...@@ -560,7 +642,18 @@ steps: ...@@ -560,7 +642,18 @@ steps:
commands: commands:
- pytest -v -s models/language/pooling -m 'not core_model' - pytest -v -s models/language/pooling -m 'not core_model'
- label: Multi-Modal Processor Test - label: Language Models Test (MTEB)
timeout_in_minutes: 110
mirror_hardwares: [amdexperimental]
optional: true
source_file_dependencies:
- vllm/
- tests/models/language/pooling_mteb_test
commands:
- pytest -v -s models/language/pooling_mteb_test
- label: Multi-Modal Processor Test # 44min
timeout_in_minutes: 60
source_file_dependencies: source_file_dependencies:
- vllm/ - vllm/
- tests/models/multimodal - tests/models/multimodal
...@@ -568,7 +661,8 @@ steps: ...@@ -568,7 +661,8 @@ steps:
- pip install git+https://github.com/TIGER-AI-Lab/Mantis.git - pip install git+https://github.com/TIGER-AI-Lab/Mantis.git
- pytest -v -s models/multimodal/processing - pytest -v -s models/multimodal/processing
- label: Multi-Modal Models Test (Standard) - label: Multi-Modal Models Test (Standard) # 60min
timeout_in_minutes: 80
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
torch_nightly: true torch_nightly: true
source_file_dependencies: source_file_dependencies:
...@@ -578,7 +672,7 @@ steps: ...@@ -578,7 +672,7 @@ steps:
- pip install git+https://github.com/TIGER-AI-Lab/Mantis.git - pip install git+https://github.com/TIGER-AI-Lab/Mantis.git
- pip freeze | grep -E 'torch' - pip freeze | grep -E 'torch'
- pytest -v -s models/multimodal -m core_model --ignore models/multimodal/generation/test_whisper.py --ignore models/multimodal/processing - pytest -v -s models/multimodal -m core_model --ignore models/multimodal/generation/test_whisper.py --ignore models/multimodal/processing
- cd .. && pytest -v -s tests/models/multimodal/generation/test_whisper.py -m core_model # Otherwise, mp_method="spawn" doesn't work - cd .. && VLLM_WORKER_MULTIPROC_METHOD=spawn pytest -v -s tests/models/multimodal/generation/test_whisper.py -m core_model # Otherwise, mp_method="spawn" doesn't work
- label: Multi-Modal Models Test (Extended) 1 - label: Multi-Modal Models Test (Extended) 1
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
...@@ -610,7 +704,8 @@ steps: ...@@ -610,7 +704,8 @@ steps:
- pip install git+https://github.com/TIGER-AI-Lab/Mantis.git - pip install git+https://github.com/TIGER-AI-Lab/Mantis.git
- pytest -v -s models/multimodal/generation/test_common.py -m 'split(group=1) and not core_model' - pytest -v -s models/multimodal/generation/test_common.py -m 'split(group=1) and not core_model'
- label: Quantized Models Test - label: Quantized Models Test # 45 min
timeout_in_minutes: 60
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
source_file_dependencies: source_file_dependencies:
- vllm/model_executor/layers/quantization - vllm/model_executor/layers/quantization
...@@ -640,7 +735,8 @@ steps: ...@@ -640,7 +735,8 @@ steps:
- python3 examples/offline_inference/audio_language.py --model-type whisper - python3 examples/offline_inference/audio_language.py --model-type whisper
- python3 examples/offline_inference/vision_language.py --model-type qwen2_5_vl - python3 examples/offline_inference/vision_language.py --model-type qwen2_5_vl
- label: Blackwell Test - label: Blackwell Test # 38 min
timeout_in_minutes: 60
working_dir: "/vllm-workspace/" working_dir: "/vllm-workspace/"
gpu: b200 gpu: b200
# optional: true # optional: true
...@@ -662,7 +758,8 @@ steps: ...@@ -662,7 +758,8 @@ steps:
# num_heads2 broken by https://github.com/flashinfer-ai/flashinfer/issues/1353 # num_heads2 broken by https://github.com/flashinfer-ai/flashinfer/issues/1353
- pytest -v -s tests/kernels/attention/test_flashinfer.py -k 'not num_heads2' - pytest -v -s tests/kernels/attention/test_flashinfer.py -k 'not num_heads2'
- pytest -v -s tests/kernels/attention/test_flashinfer_trtllm_attention.py - pytest -v -s tests/kernels/attention/test_flashinfer_trtllm_attention.py
- pytest -v -s tests/kernels/test_cutlass_mla_decode.py - pytest -v -s tests/kernels/attention/test_cutlass_mla_decode.py
- pytest -v -s tests/kernels/attention/test_flashinfer_mla_decode.py
# Quantization # Quantization
- pytest -v -s tests/kernels/quantization/test_cutlass_scaled_mm.py -k 'fp8' - pytest -v -s tests/kernels/quantization/test_cutlass_scaled_mm.py -k 'fp8'
- pytest -v -s tests/kernels/quantization/test_nvfp4_quant.py - pytest -v -s tests/kernels/quantization/test_nvfp4_quant.py
...@@ -682,6 +779,7 @@ steps: ...@@ -682,6 +779,7 @@ steps:
##### multi gpus test ##### ##### multi gpus test #####
- label: Distributed Comm Ops Test # 7min - label: Distributed Comm Ops Test # 7min
timeout_in_minutes: 20
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
working_dir: "/vllm-workspace/tests" working_dir: "/vllm-workspace/tests"
num_gpus: 2 num_gpus: 2
...@@ -693,6 +791,7 @@ steps: ...@@ -693,6 +791,7 @@ steps:
- pytest -v -s distributed/test_shm_broadcast.py - pytest -v -s distributed/test_shm_broadcast.py
- label: 2 Node Tests (4 GPUs in total) # 16min - label: 2 Node Tests (4 GPUs in total) # 16min
timeout_in_minutes: 30
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
working_dir: "/vllm-workspace/tests" working_dir: "/vllm-workspace/tests"
num_gpus: 2 num_gpus: 2
...@@ -716,7 +815,8 @@ steps: ...@@ -716,7 +815,8 @@ steps:
- NUM_NODES=2 torchrun --nnodes 2 --nproc-per-node=2 --rdzv_backend=c10d --rdzv_endpoint=192.168.10.10 distributed/test_node_count.py | grep 'Node count test passed' - NUM_NODES=2 torchrun --nnodes 2 --nproc-per-node=2 --rdzv_backend=c10d --rdzv_endpoint=192.168.10.10 distributed/test_node_count.py | grep 'Node count test passed'
- python3 ../examples/offline_inference/data_parallel.py --dp-size=2 --tp-size=1 --node-size=2 --node-rank=1 --master-addr=192.168.10.10 --master-port=12345 --enforce-eager --trust-remote-code - python3 ../examples/offline_inference/data_parallel.py --dp-size=2 --tp-size=1 --node-size=2 --node-rank=1 --master-addr=192.168.10.10 --master-port=12345 --enforce-eager --trust-remote-code
- label: Distributed Tests (2 GPUs) # 40min - label: Distributed Tests (2 GPUs) # 110min
timeout_in_minutes: 150
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
working_dir: "/vllm-workspace/tests" working_dir: "/vllm-workspace/tests"
num_gpus: 2 num_gpus: 2
...@@ -747,7 +847,8 @@ steps: ...@@ -747,7 +847,8 @@ steps:
# Avoid importing model tests that cause CUDA reinitialization error # Avoid importing model tests that cause CUDA reinitialization error
- pytest models/test_transformers.py -v -s -m 'distributed(num_gpus=2)' - pytest models/test_transformers.py -v -s -m 'distributed(num_gpus=2)'
- pytest models/language -v -s -m 'distributed(num_gpus=2)' - pytest models/language -v -s -m 'distributed(num_gpus=2)'
- pytest models/multimodal -v -s -m 'distributed(num_gpus=2)' - pytest models/multimodal -v -s -m 'distributed(num_gpus=2)' --ignore models/multimodal/generation/test_whisper.py
- VLLM_WORKER_MULTIPROC_METHOD=spawn pytest models/multimodal/generation/test_whisper.py -v -s -m 'distributed(num_gpus=2)'
# test sequence parallel # test sequence parallel
- pytest -v -s distributed/test_sequence_parallel.py - pytest -v -s distributed/test_sequence_parallel.py
# this test fails consistently. # this test fails consistently.
...@@ -757,6 +858,7 @@ steps: ...@@ -757,6 +858,7 @@ steps:
- pytest -v -s models/multimodal/generation/test_maverick.py - pytest -v -s models/multimodal/generation/test_maverick.py
- label: Plugin Tests (2 GPUs) # 40min - label: Plugin Tests (2 GPUs) # 40min
timeout_in_minutes: 60
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
working_dir: "/vllm-workspace/tests" working_dir: "/vllm-workspace/tests"
num_gpus: 2 num_gpus: 2
...@@ -772,7 +874,7 @@ steps: ...@@ -772,7 +874,7 @@ steps:
# begin io_processor plugins test, all the code in between uses the prithvi_io_processor plugin # begin io_processor plugins test, all the code in between uses the prithvi_io_processor plugin
- pip install -e ./plugins/prithvi_io_processor_plugin - pip install -e ./plugins/prithvi_io_processor_plugin
- pytest -v -s plugins_tests/test_io_processor_plugins.py - pytest -v -s plugins_tests/test_io_processor_plugins.py
- pip uninstall prithvi_io_processor_plugin -y - pip uninstall prithvi_io_processor_plugin -y
# end io_processor plugins test # end io_processor plugins test
# other tests continue here: # other tests continue here:
- pytest -v -s plugins_tests/test_scheduler_plugins.py - pytest -v -s plugins_tests/test_scheduler_plugins.py
...@@ -782,7 +884,8 @@ steps: ...@@ -782,7 +884,8 @@ steps:
- pytest -v -s models/test_oot_registration.py # it needs a clean process - pytest -v -s models/test_oot_registration.py # it needs a clean process
- pytest -v -s plugins/lora_resolvers # unit tests for in-tree lora resolver plugins - pytest -v -s plugins/lora_resolvers # unit tests for in-tree lora resolver plugins
- label: Pipeline Parallelism Test # 45min - label: Pipeline + Context Parallelism Test # 45min
timeout_in_minutes: 60
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
working_dir: "/vllm-workspace/tests" working_dir: "/vllm-workspace/tests"
num_gpus: 4 num_gpus: 4
...@@ -795,8 +898,10 @@ steps: ...@@ -795,8 +898,10 @@ steps:
commands: commands:
- pytest -v -s distributed/test_pp_cudagraph.py - pytest -v -s distributed/test_pp_cudagraph.py
- pytest -v -s distributed/test_pipeline_parallel.py - pytest -v -s distributed/test_pipeline_parallel.py
# - pytest -v -s distributed/test_context_parallel.py # TODO: enable it on Hopper runners or add triton MLA support
- label: LoRA TP Test (Distributed) - label: LoRA TP Test (Distributed) # 17 min
timeout_in_minutes: 30
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
num_gpus: 4 num_gpus: 4
source_file_dependencies: source_file_dependencies:
...@@ -814,9 +919,10 @@ steps: ...@@ -814,9 +919,10 @@ steps:
- label: Weight Loading Multiple GPU Test # 33min - label: Weight Loading Multiple GPU Test # 33min
timeout_in_minutes: 45
mirror_hardwares: [amdexperimental] mirror_hardwares: [amdexperimental]
working_dir: "/vllm-workspace/tests" working_dir: "/vllm-workspace/tests"
num_gpus: 2 num_gpus: 2
optional: true optional: true
source_file_dependencies: source_file_dependencies:
- vllm/ - vllm/
......
.github/.bc-linter.yml 0 → 100644
View file @ 38d80967
# doc: https://github.com/pytorch/test-infra/blob/main/tools/stronghold/docs/bc_linter_config.md
version: 1
paths:
# We temporarily disable globally, and will only enable with `annotations.include`
# include:
# - "vllm/v1/attetion/*.py"
# - "vllm/v1/core/*.py"
exclude:
- "**/*.py"
scan:
functions: true # check free functions and methods
classes: true # check classes/dataclasses
public_only: true # ignore names starting with "_" at any level
annotations:
include: # decorators that force‑include a symbol
- name: "bc_linter_include" # matched by simple name or dotted suffix
propagate_to_members: false # for classes, include methods/inner classes
exclude: # decorators that force‑exclude a symbol
- name: "bc_linter_skip" # matched by simple name or dotted suffix
propagate_to_members: true # for classes, exclude methods/inner classes
excluded_violations: [] # e.g. ["ParameterRenamed", "FieldTypeChanged"]
.github/CODEOWNERS
View file @ 38d80967
...@@ -5,18 +5,21 @@ ...@@ -5,18 +5,21 @@
/vllm/attention/backends/abstract.py @WoosukKwon @zhuohan123 @youkaichao @alexm-redhat @comaniac @njhill /vllm/attention/backends/abstract.py @WoosukKwon @zhuohan123 @youkaichao @alexm-redhat @comaniac @njhill
/vllm/core @zhuohan123 @youkaichao @alexm-redhat @comaniac @njhill /vllm/core @zhuohan123 @youkaichao @alexm-redhat @comaniac @njhill
/vllm/engine/llm_engine.py @zhuohan123 @youkaichao @alexm-redhat @comaniac @njhill /vllm/engine/llm_engine.py @zhuohan123 @youkaichao @alexm-redhat @comaniac @njhill
/vllm/executor/executor_base.py @zhuohan123 @youkaichao @alexm-redhat @comaniac @njhill /vllm/executor/executor_base.py @zhuohan123 @youkaichao @alexm-redhat @comaniac @njhill @22quinn
/vllm/worker/worker_base.py @zhuohan123 @youkaichao @alexm-redhat @comaniac @njhill /vllm/worker/worker_base.py @zhuohan123 @youkaichao @alexm-redhat @comaniac @njhill @22quinn
/vllm/worker/worker.py @zhuohan123 @youkaichao @alexm-redhat @comaniac @njhill /vllm/worker/worker.py @zhuohan123 @youkaichao @alexm-redhat @comaniac @njhill
/vllm/model_executor/layers/sampler.py @zhuohan123 @youkaichao @alexm-redhat @comaniac @njhill /vllm/model_executor/layers/sampler.py @zhuohan123 @youkaichao @alexm-redhat @comaniac @njhill @NickLucche
/vllm/model_executor/layers/quantization @mgoin @robertgshaw2-redhat @tlrmchlsmth @yewentao256 /vllm/model_executor/layers/quantization @mgoin @robertgshaw2-redhat @tlrmchlsmth @yewentao256
/vllm/model_executor/layers/mamba @tdoublep /vllm/model_executor/layers/mamba @tdoublep
/vllm/multimodal @DarkLight1337 @ywang96 /vllm/model_executor/model_loader @22quinn
/vllm/multimodal @DarkLight1337 @ywang96 @NickLucche
/vllm/v1/sample @22quinn @houseroad
/vllm/vllm_flash_attn @LucasWilkinson /vllm/vllm_flash_attn @LucasWilkinson
/vllm/lora @jeejeelee /vllm/lora @jeejeelee
/vllm/reasoning @aarnphm /vllm/reasoning @aarnphm @chaunceyjiang
/vllm/entrypoints @aarnphm /vllm/entrypoints @aarnphm @chaunceyjiang
/vllm/compilation @zou3519 @youkaichao @ProExpertProg /vllm/compilation @zou3519 @youkaichao @ProExpertProg
/vllm/distributed/kv_transfer @NickLucche
CMakeLists.txt @tlrmchlsmth @LucasWilkinson CMakeLists.txt @tlrmchlsmth @LucasWilkinson
# Any change to the VllmConfig changes can have a large user-facing impact, # Any change to the VllmConfig changes can have a large user-facing impact,
...@@ -25,8 +28,11 @@ CMakeLists.txt @tlrmchlsmth @LucasWilkinson ...@@ -25,8 +28,11 @@ CMakeLists.txt @tlrmchlsmth @LucasWilkinson
# vLLM V1 # vLLM V1
/vllm/v1 @WoosukKwon @robertgshaw2-redhat @njhill @ywang96 @comaniac @alexm-redhat /vllm/v1 @WoosukKwon @robertgshaw2-redhat @njhill @ywang96 @comaniac @alexm-redhat
/vllm/v1/structured_output @mgoin @russellb @aarnphm /vllm/v1/structured_output @mgoin @russellb @aarnphm @benchislett
/vllm/v1/spec_decode @benchislett @luccafong
/vllm/v1/attention/backends/triton_attn.py @tdoublep /vllm/v1/attention/backends/triton_attn.py @tdoublep
/vllm/v1/core @heheda12345
/vllm/v1/kv_cache_interface.py @heheda12345
# Test ownership # Test ownership
/.buildkite/lm-eval-harness @mgoin @simon-mo /.buildkite/lm-eval-harness @mgoin @simon-mo
...@@ -34,18 +40,20 @@ CMakeLists.txt @tlrmchlsmth @LucasWilkinson ...@@ -34,18 +40,20 @@ CMakeLists.txt @tlrmchlsmth @LucasWilkinson
/tests/distributed/test_multi_node_assignment.py @youkaichao /tests/distributed/test_multi_node_assignment.py @youkaichao
/tests/distributed/test_pipeline_parallel.py @youkaichao /tests/distributed/test_pipeline_parallel.py @youkaichao
/tests/distributed/test_same_node.py @youkaichao /tests/distributed/test_same_node.py @youkaichao
/tests/entrypoints @DarkLight1337 @robertgshaw2-redhat @simon-mo @aarnphm /tests/entrypoints @DarkLight1337 @robertgshaw2-redhat @simon-mo @aarnphm @NickLucche
/tests/kernels @tlrmchlsmth @WoosukKwon @yewentao256 /tests/kernels @tlrmchlsmth @WoosukKwon @yewentao256
/tests/models @DarkLight1337 @ywang96 /tests/models @DarkLight1337 @ywang96
/tests/multimodal @DarkLight1337 @ywang96 /tests/multimodal @DarkLight1337 @ywang96 @NickLucche
/tests/prefix_caching @comaniac @KuntaiDu /tests/prefix_caching @comaniac @KuntaiDu
/tests/quantization @mgoin @robertgshaw2-redhat @yewentao256 /tests/quantization @mgoin @robertgshaw2-redhat @yewentao256
/tests/test_inputs.py @DarkLight1337 @ywang96 /tests/test_inputs.py @DarkLight1337 @ywang96
/tests/v1/entrypoints/llm/test_struct_output_generate.py @mgoin @russellb @aarnphm /tests/v1/entrypoints/llm/test_struct_output_generate.py @mgoin @russellb @aarnphm
/tests/v1/structured_output @mgoin @russellb @aarnphm /tests/v1/structured_output @mgoin @russellb @aarnphm
/tests/v1/core @heheda12345
/tests/weight_loading @mgoin @youkaichao @yewentao256 /tests/weight_loading @mgoin @youkaichao @yewentao256
/tests/lora @jeejeelee /tests/lora @jeejeelee
/tests/models/language/generation/test_hybrid.py @tdoublep /tests/models/language/generation/test_hybrid.py @tdoublep
/tests/v1/kv_connector/nixl_integration @NickLucche
# Docs # Docs
/docs @hmellor /docs @hmellor
...@@ -67,6 +75,9 @@ mkdocs.yaml @hmellor ...@@ -67,6 +75,9 @@ mkdocs.yaml @hmellor
/vllm/attention/backends/dual_chunk_flash_attn.py @sighingnow /vllm/attention/backends/dual_chunk_flash_attn.py @sighingnow
/vllm/model_executor/models/qwen* @sighingnow /vllm/model_executor/models/qwen* @sighingnow
# MTP-specific files
/vllm/model_executor/models/deepseek_mtp.py @luccafong
# Mistral-specific files # Mistral-specific files
/vllm/model_executor/models/mistral*.py @patrickvonplaten /vllm/model_executor/models/mistral*.py @patrickvonplaten
/vllm/model_executor/models/mixtral*.py @patrickvonplaten /vllm/model_executor/models/mixtral*.py @patrickvonplaten
...@@ -86,3 +97,8 @@ mkdocs.yaml @hmellor ...@@ -86,3 +97,8 @@ mkdocs.yaml @hmellor
/vllm/attention/ops/rocm*.py @gshtras /vllm/attention/ops/rocm*.py @gshtras
/vllm/model_executor/layers/fused_moe/rocm*.py @gshtras /vllm/model_executor/layers/fused_moe/rocm*.py @gshtras
# TPU
/vllm/v1/worker/tpu* @NickLucche
/vllm/platforms/tpu.py @NickLucche
/vllm/v1/sample/tpu @NickLucche
/vllm/tests/v1/tpu @NickLucche
\ No newline at end of file
.github/mergify.yml
View file @ 38d80967
...@@ -273,6 +273,20 @@ pull_request_rules: ...@@ -273,6 +273,20 @@ pull_request_rules:
users: users:
- "sangstar" - "sangstar"
- name: assign reviewer for modelopt changes
conditions:
- or:
- files~=^vllm/model_executor/layers/quantization/modelopt\.py$
- files~=^vllm/model_executor/layers/quantization/__init__\.py$
- files~=^tests/models/quantization/test_modelopt\.py$
- files~=^tests/quantization/test_modelopt\.py$
- files~=^tests/models/quantization/test_nvfp4\.py$
- files~=^docs/features/quantization/modelopt\.md$
actions:
assign:
users:
- "Edwardf0t1"
- name: remove 'needs-rebase' label when conflict is resolved - name: remove 'needs-rebase' label when conflict is resolved
conditions: conditions:
- -conflict - -conflict
......
.github/workflows/add_label_automerge.yml
View file @ 38d80967
...@@ -10,7 +10,7 @@ jobs: ...@@ -10,7 +10,7 @@ jobs:
runs-on: ubuntu-latest runs-on: ubuntu-latest
steps: steps:
- name: Add label - name: Add label
uses: actions/github-script@60a0d83039c74a4aee543508d2ffcb1c3799cdea # v7.0.1 uses: actions/github-script@ed597411d8f924073f98dfc5c65a23a2325f34cd # v8.0.0
with: with:
script: | script: |
github.rest.issues.addLabels({ github.rest.issues.addLabels({
......
.github/workflows/bc-lint.yml 0 → 100644
View file @ 38d80967
name: BC Lint
on:
pull_request:
types:
- opened
- synchronize
- reopened
jobs:
bc_lint:
if: github.repository_owner == 'vllm-project'
runs-on: ubuntu-latest
steps:
- name: Run BC Lint Action
uses: pytorch/test-infra/.github/actions/bc-lint@main
with:
repo: ${{ github.event.pull_request.head.repo.full_name }}
base_sha: ${{ github.event.pull_request.base.sha }}
head_sha: ${{ github.event.pull_request.head.sha }}
suppression: ${{ contains(github.event.pull_request.labels.*.name, 'suppress-bc-linter') }}
docs_link: 'https://github.com/pytorch/test-infra/wiki/BC-Linter'
config_dir: .github
concurrency:
group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.sha }}
cancel-in-progress: true
.github/workflows/cleanup_pr_body.yml
View file @ 38d80967
...@@ -16,7 +16,7 @@ jobs: ...@@ -16,7 +16,7 @@ jobs:
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2 uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
- name: Set up Python - name: Set up Python
uses: actions/setup-python@42375524e23c412d93fb67b49958b491fce71c38 # v5.4.0 uses: actions/setup-python@e797f83bcb11b83ae66e0230d6156d7c80228e7c # v6.0.0
with: with:
python-version: '3.12' python-version: '3.12'
......
.github/workflows/issue_autolabel.yml
View file @ 38d80967
...@@ -13,7 +13,7 @@ jobs: ...@@ -13,7 +13,7 @@ jobs:
runs-on: ubuntu-latest runs-on: ubuntu-latest
steps: steps:
- name: Label issues based on keywords - name: Label issues based on keywords
uses: actions/github-script@60a0d83039c74a4aee543508d2ffcb1c3799cdea # v7.0.1 uses: actions/github-script@ed597411d8f924073f98dfc5c65a23a2325f34cd # v8.0.0
with: with:
script: | script: |
// Configuration: Add new labels and keywords here // Configuration: Add new labels and keywords here
......
.github/workflows/pre-commit.yml
View file @ 38d80967
...@@ -17,7 +17,7 @@ jobs: ...@@ -17,7 +17,7 @@ jobs:
runs-on: ubuntu-latest runs-on: ubuntu-latest
steps: steps:
- uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2 - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
- uses: actions/setup-python@42375524e23c412d93fb67b49958b491fce71c38 # v5.4.0 - uses: actions/setup-python@e797f83bcb11b83ae66e0230d6156d7c80228e7c # v6.0.0
with: with:
python-version: "3.12" python-version: "3.12"
- run: echo "::add-matcher::.github/workflows/matchers/actionlint.json" - run: echo "::add-matcher::.github/workflows/matchers/actionlint.json"
......
.github/workflows/reminder_comment.yml
View file @ 38d80967
...@@ -9,7 +9,7 @@ jobs: ...@@ -9,7 +9,7 @@ jobs:
runs-on: ubuntu-latest runs-on: ubuntu-latest
steps: steps:
- name: Remind to run full CI on PR - name: Remind to run full CI on PR
uses: actions/github-script@60a0d83039c74a4aee543508d2ffcb1c3799cdea # v7.0.1 uses: actions/github-script@ed597411d8f924073f98dfc5c65a23a2325f34cd # v8.0.0
with: with:
script: | script: |
try { try {
......
.github/workflows/stale.yml
View file @ 38d80967
...@@ -13,7 +13,7 @@ jobs: ...@@ -13,7 +13,7 @@ jobs:
actions: write actions: write
runs-on: ubuntu-latest runs-on: ubuntu-latest
steps: steps:
- uses: actions/stale@5bef64f19d7facfb25b37b414482c7164d639639 # v9.1.0 - uses: actions/stale@3a9db7e6a41a89f618792c92c0e97cc736e1b13f # v10.0.0
with: with:
# Increasing this value ensures that changes to this workflow # Increasing this value ensures that changes to this workflow
# propagate to all issues and PRs in days rather than months # propagate to all issues and PRs in days rather than months
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment