[Misc] Add SPDX-FileCopyrightText (#19100)

Signed-off-by: simon-mo <simon.mo@hey.com>
This commit is contained in:
Simon Mo
2025-06-03 11:20:17 -07:00
committed by GitHub
parent d054da1992
commit 02f0c7b220
1432 changed files with 1441 additions and 6 deletions
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import os import os
import sys import sys
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import argparse import argparse
import os import os
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
from pathlib import Path from pathlib import Path
import pytest import pytest
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
""" """
LM eval harness on model to compare vs HF baseline computed offline. LM eval harness on model to compare vs HF baseline computed offline.
Configs are found in configs/$MODEL.yaml Configs are found in configs/$MODEL.yaml
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import json import json
import os import os
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import argparse import argparse
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import argparse import argparse
import json import json
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
from lmdeploy.serve.openai.api_client import APIClient from lmdeploy.serve.openai.api_client import APIClient
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import datetime import datetime
import json import json
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import io import io
import json import json
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
""" """
This module defines a framework for sampling benchmark requests from various This module defines a framework for sampling benchmark requests from various
datasets. Each dataset subclass of BenchmarkDataset must implement sample datasets. Each dataset subclass of BenchmarkDataset must implement sample
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
"""Benchmark the latency of processing a single batch of requests.""" """Benchmark the latency of processing a single batch of requests."""
import argparse import argparse
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
""" """
Offline benchmark to test the long document QA throughput. Offline benchmark to test the long document QA throughput.
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
""" """
Benchmark the efficiency of prefix caching. Benchmark the efficiency of prefix caching.
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
"""Benchmark offline prioritization.""" """Benchmark offline prioritization."""
import argparse import argparse
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
r"""Benchmark online serving throughput. r"""Benchmark online serving throughput.
On the server side, run one of the following commands: On the server side, run one of the following commands:
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
r"""Benchmark online serving throughput with structured outputs. r"""Benchmark online serving throughput with structured outputs.
On the server side, run one of the following commands: On the server side, run one of the following commands:
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
"""Benchmark offline inference throughput.""" """Benchmark offline inference throughput."""
import argparse import argparse
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import argparse import argparse
import json import json
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import argparse import argparse
import copy import copy
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
# Cutlass bench utils # Cutlass bench utils
from collections.abc import Iterable from collections.abc import Iterable
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import argparse import argparse
import copy import copy
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
# Weight Shapes are in the format # Weight Shapes are in the format
# ([K, N], TP_SPLIT_DIM) # ([K, N], TP_SPLIT_DIM)
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import os import os
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import asyncio import asyncio
import itertools import itertools
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import json import json
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import pickle as pkl import pickle as pkl
import time import time
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import argparse import argparse
import copy import copy
import itertools import itertools
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import os import os
import sys import sys
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
# Copyright (c) Microsoft Corporation. # Copyright (c) Microsoft Corporation.
# Licensed under the MIT License. # Licensed under the MIT License.
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
""" """
Benchmark the performance of the cutlass_moe_fp4 kernel vs the triton_moe Benchmark the performance of the cutlass_moe_fp4 kernel vs the triton_moe
kernel. The cutlass_moe_fp4 kernel takes in fp4 quantized weights and 16-bit kernel. The cutlass_moe_fp4 kernel takes in fp4 quantized weights and 16-bit
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import torch import torch
import torch.utils.benchmark as benchmark import torch.utils.benchmark as benchmark
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import time import time
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import argparse import argparse
import copy import copy
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import argparse import argparse
import copy import copy
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import torch import torch
import torch.utils.benchmark as benchmark import torch.utils.benchmark as benchmark
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import argparse import argparse
import json import json
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import argparse import argparse
from typing import Any, TypedDict from typing import Any, TypedDict
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import random import random
import time import time
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import time import time
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import itertools import itertools
from typing import Optional, Union from typing import Optional, Union
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
from itertools import accumulate from itertools import accumulate
from typing import Optional from typing import Optional
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
WEIGHT_SHAPES = { WEIGHT_SHAPES = {
"ideal": [[4 * 256 * 32, 256 * 32]], "ideal": [[4 * 256 * 32, 256 * 32]],
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
# Adapted from sglang quantization/tuning_block_wise_kernel.py # Adapted from sglang quantization/tuning_block_wise_kernel.py
import argparse import argparse
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
# fmt: off # fmt: off
# ruff: noqa: E501 # ruff: noqa: E501
import time import time
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import math import math
import pickle import pickle
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import dataclasses import dataclasses
from collections.abc import Iterable from collections.abc import Iterable
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
# Weight Shapes are in the format # Weight Shapes are in the format
# ([K, N], TP_SPLIT_DIM) # ([K, N], TP_SPLIT_DIM)
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import cProfile import cProfile
import pstats import pstats
+1
View File
@@ -1,5 +1,6 @@
#!/usr/bin/env python3 #!/usr/bin/env python3
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
# #
# A command line tool for running pytorch's hipify preprocessor on CUDA # A command line tool for running pytorch's hipify preprocessor on CUDA
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import enum import enum
from typing import Union from typing import Union
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import glob import glob
import itertools import itertools
import os import os
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import glob import glob
import itertools import itertools
import os import os
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import itertools import itertools
import math import math
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import itertools import itertools
from dataclasses import dataclass, field from dataclasses import dataclass, field
from pathlib import Path from pathlib import Path
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import os import os
from typing import Literal from typing import Literal
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import regex as re import regex as re
from mkdocs.config.defaults import MkDocsConfig from mkdocs.config.defaults import MkDocsConfig
from mkdocs.structure.files import Files from mkdocs.structure.files import Files
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
""" """
This example shows how to use vLLM for running offline inference This example shows how to use vLLM for running offline inference
with the correct prompt format on audio language models. with the correct prompt format on audio language models.
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
""" """
Demonstration script for Automatic Prefix Caching (APC) in vLLM. Demonstration script for Automatic Prefix Caching (APC) in vLLM.
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
from vllm import LLM, SamplingParams from vllm import LLM, SamplingParams
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
from vllm import LLM, EngineArgs from vllm import LLM, EngineArgs
from vllm.utils import FlexibleArgumentParser from vllm.utils import FlexibleArgumentParser
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
from argparse import Namespace from argparse import Namespace
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
from argparse import Namespace from argparse import Namespace
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
from vllm import LLM, EngineArgs from vllm import LLM, EngineArgs
from vllm.utils import FlexibleArgumentParser from vllm.utils import FlexibleArgumentParser
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
from argparse import Namespace from argparse import Namespace
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
""" """
This example shows how to use Ray Data for data parallel batch inference. This example shows how to use Ray Data for data parallel batch inference.
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
# ruff: noqa # ruff: noqa
import json import json
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
from vllm import LLM, SamplingParams from vllm import LLM, SamplingParams
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
""" """
Usage: Usage:
Single node: Single node:
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
from vllm import LLM, SamplingParams from vllm import LLM, SamplingParams
from vllm.config import KVTransferConfig from vllm.config import KVTransferConfig
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
from vllm import LLM, SamplingParams from vllm import LLM, SamplingParams
from vllm.config import KVTransferConfig from vllm.config import KVTransferConfig
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
""" """
This file demonstrates the example usage of disaggregated prefilling This file demonstrates the example usage of disaggregated prefilling
We will launch 2 vllm instances (GPU 0 for prefill and GPU 1 for decode), We will launch 2 vllm instances (GPU 0 for prefill and GPU 1 for decode),
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import argparse import argparse
import json import json
import os import os
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
from argparse import Namespace from argparse import Namespace
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
from argparse import Namespace from argparse import Namespace
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
""" """
Demonstrate prompting of text-to-text Demonstrate prompting of text-to-text
encoder/decoder models, specifically BART encoder/decoder models, specifically BART
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
""" """
This example shows how to use vLLM for running offline inference with This example shows how to use vLLM for running offline inference with
the explicit/implicit prompt format on enc-dec LMMs for text generation. the explicit/implicit prompt format on enc-dec LMMs for text generation.
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
""" """
This file demonstrates using the `LLMEngine` This file demonstrates using the `LLMEngine`
for processing prompts with various sampling parameters. for processing prompts with various sampling parameters.
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
""" """
Validates the loading of a model saved with the sharded_state format. Validates the loading of a model saved with the sharded_state format.
This script demonstrates how to load a model that was previously saved This script demonstrates how to load a model that was previously saved
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
""" """
This example shows how to use LoRA with different quantization techniques This example shows how to use LoRA with different quantization techniques
for offline inference. for offline inference.
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
from vllm import LLM, SamplingParams from vllm import LLM, SamplingParams
from vllm.v1.metrics.reader import Counter, Gauge, Histogram, Vector from vllm.v1.metrics.reader import Counter, Gauge, Histogram, Vector
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
# ruff: noqa # ruff: noqa
import argparse import argparse
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
""" """
This file demonstrates the usage of text generation with an LLM model, This file demonstrates the usage of text generation with an LLM model,
comparing the performance with and without speculative decoding. comparing the performance with and without speculative decoding.
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
""" """
This example shows how to use the multi-LoRA functionality This example shows how to use the multi-LoRA functionality
for offline inference. for offline inference.
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
from vllm import LLM, SamplingParams from vllm import LLM, SamplingParams
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
""" """
This example shows how to run offline inference with an EAGLE speculative This example shows how to run offline inference with an EAGLE speculative
decoding model on neuron. To use EAGLE speculative decoding, you must use decoding model on neuron. To use EAGLE speculative decoding, you must use
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import os import os
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import requests import requests
import torch import torch
from neuronx_distributed_inference.models.mllama.utils import add_instruct from neuronx_distributed_inference.models.mllama.utils import add_instruct
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
""" """
This example shows how to run offline inference with a speculative This example shows how to run offline inference with a speculative
decoding model on neuron. decoding model on neuron.
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
from vllm import LLM, SamplingParams from vllm import LLM, SamplingParams
from vllm.distributed import cleanup_dist_env_and_memory from vllm.distributed import cleanup_dist_env_and_memory
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
""" """
This is a demo script showing how to use the This is a demo script showing how to use the
PrithviGeospatialMAE model with vLLM PrithviGeospatialMAE model with vLLM
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import inspect import inspect
import json import json
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import argparse import argparse
import dataclasses import dataclasses
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
""" """
Demonstrates how to generate prompt embeddings using Demonstrates how to generate prompt embeddings using
Hugging Face Transformers and use them as input to vLLM Hugging Face Transformers and use them as input to vLLM
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
""" """
This example shows how to use vLLM for running offline inference This example shows how to use vLLM for running offline inference
with the correct prompt format on Qwen2.5-Omni (thinker only). with the correct prompt format on Qwen2.5-Omni (thinker only).
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import os import os
from urllib.request import urlopen from urllib.request import urlopen
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
""" """
Demonstrates how to achieve reproducibility in vLLM. Demonstrates how to achieve reproducibility in vLLM.
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
""" """
a simple demonstration of RLHF with vLLM, inspired by a simple demonstration of RLHF with vLLM, inspired by
the OpenRLHF framework https://github.com/OpenRLHF/OpenRLHF . the OpenRLHF framework https://github.com/OpenRLHF/OpenRLHF .
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
""" """
a simple demonstration to show how to co-locate a simple demonstration to show how to co-locate
vLLM worker with training actors on the same GPUs, vLLM worker with training actors on the same GPUs,
+1
View File
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import torch import torch

Some files were not shown because too many files have changed in this diff Show More