[Misc] Add SPDX-License-Identifier headers to python source files (#12628)

- **Add SPDX license headers to python source files**
- **Check for SPDX headers using pre-commit**

commit 9d7ef44c3cfb72ca4c32e1c677d99259d10d4745
Author: Russell Bryant <rbryant@redhat.com>
Date:   Fri Jan 31 14:18:24 2025 -0500

    Add SPDX license headers to python source files
    
This commit adds SPDX license headers to python source files as
recommended to
the project by the Linux Foundation. These headers provide a concise way
that is
both human and machine readable for communicating license information
for each
source file. It helps avoid any ambiguity about the license of the code
and can
    also be easily used by tools to help manage license compliance.
    
The Linux Foundation runs license scans against the codebase to help
ensure
    we are in compliance with the licenses of the code we use, including
dependencies. Having these headers in place helps that tool do its job.
    
    More information can be found on the SPDX site:
    
    - https://spdx.dev/learn/handling-license-info/
    
    Signed-off-by: Russell Bryant <rbryant@redhat.com>

commit 5a1cf1cb3b80759131c73f6a9dddebccac039dea
Author: Russell Bryant <rbryant@redhat.com>
Date:   Fri Jan 31 14:36:32 2025 -0500

    Check for SPDX headers using pre-commit
    
    Signed-off-by: Russell Bryant <rbryant@redhat.com>

---------

Signed-off-by: Russell Bryant <rbryant@redhat.com>

.buildkite/check-wheel-size.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import os
 import sys
 import zipfile

.buildkite/generate_index.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import argparse
 import os
 

.buildkite/lm-eval-harness/test_lm_eval_correctness.py

@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: Apache-2.0
 """
 LM eval harness on model to compare vs HF baseline computed offline.
 Configs are found in configs/$MODEL.yaml

.buildkite/nightly-benchmarks/scripts/convert-results-json-to-markdown.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import json
 import os
 from pathlib import Path

.buildkite/nightly-benchmarks/scripts/download-tokenizer.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import argparse
 
 from transformers import AutoTokenizer

.buildkite/nightly-benchmarks/scripts/generate-nightly-markdown.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import argparse
 import json
 from pathlib import Path

.buildkite/nightly-benchmarks/scripts/get-lmdeploy-modelname.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 from lmdeploy.serve.openai.api_client import APIClient
 
 api_client = APIClient("http://localhost:8000")

.buildkite/nightly-benchmarks/scripts/summary-nightly-results.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import datetime
 import json
 import os

.pre-commit-config.yaml

@@ -97,10 +97,14 @@ repos:
     language: system
     verbose: true
     stages: [commit-msg]
+  - id: check-spdx-header
+    name: Check SPDX headers
+    entry: python tools/check_spdx_header.py
+    language: python
+    types: [python]
   - id: suggestion
     name: Suggestion
     entry: bash -c 'echo "To bypass pre-commit hooks, add --no-verify to git commit."'
     language: system
     verbose: true
     pass_filenames: false
-

benchmarks/backend_request_func.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import json
 import os
 import sys

benchmarks/benchmark_guided.py

@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: Apache-2.0
 """Benchmark guided decoding throughput."""
 import argparse
 import dataclasses

benchmarks/benchmark_latency.py

@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: Apache-2.0
 """Benchmark the latency of processing a single batch of requests."""
 import argparse
 import dataclasses

benchmarks/benchmark_long_document_qa_throughput.py

@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: Apache-2.0
 """
 Offline benchmark to test the long document QA throughput.
 

benchmarks/benchmark_prefix_caching.py

@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: Apache-2.0
 """
 Benchmark the efficiency of prefix caching.
 

benchmarks/benchmark_prioritization.py

@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: Apache-2.0
 """Benchmark offline prioritization."""
 import argparse
 import dataclasses

benchmarks/benchmark_serving.py

@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: Apache-2.0
 r"""Benchmark online serving throughput.
 
 On the server side, run one of the following commands:

benchmarks/benchmark_serving_guided.py

@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: Apache-2.0
 r"""Benchmark online serving throughput with guided decoding.
 
 On the server side, run one of the following commands:

benchmarks/benchmark_throughput.py

@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: Apache-2.0
 """Benchmark offline inference throughput."""
 import argparse
 import dataclasses

benchmarks/cutlass_benchmarks/sparse_benchmarks.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import argparse
 import copy
 import itertools

benchmarks/cutlass_benchmarks/utils.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 # Cutlass bench utils
 from typing import Iterable, Tuple
 

benchmarks/cutlass_benchmarks/w8a8_benchmarks.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import argparse
 import copy
 import itertools

benchmarks/cutlass_benchmarks/weight_shapes.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 # Weight Shapes are in the format
 # ([K, N], TP_SPLIT_DIM)
 # Example:

benchmarks/disagg_benchmarks/disagg_prefill_proxy_server.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import os
 
 import aiohttp

benchmarks/disagg_benchmarks/round_robin_proxy.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import asyncio
 import itertools
 

benchmarks/disagg_benchmarks/visualize_benchmark_results.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import json
 
 import matplotlib.pyplot as plt

benchmarks/fused_kernels/layernorm_rms_benchmarks.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import pickle as pkl
 import time
 from dataclasses import dataclass

benchmarks/kernels/benchmark_aqlm.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import os
 import sys
 from typing import Optional

benchmarks/kernels/benchmark_layernorm.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import time
 
 import torch

benchmarks/kernels/benchmark_lora.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import argparse
 import copy
 import json

benchmarks/kernels/benchmark_machete.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import argparse
 import copy
 import itertools

benchmarks/kernels/benchmark_marlin.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 from typing import List
 
 import torch

benchmarks/kernels/benchmark_moe.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import argparse
 import time
 from datetime import datetime

benchmarks/kernels/benchmark_paged_attention.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import random
 import time
 from typing import List, Optional

benchmarks/kernels/benchmark_quant.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import time
 
 import torch

benchmarks/kernels/benchmark_rmsnorm.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import itertools
 from typing import Optional, Tuple, Union
 

benchmarks/kernels/benchmark_rope.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 from itertools import accumulate
 from typing import List, Optional
 

benchmarks/kernels/benchmark_shapes.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 WEIGHT_SHAPES = {
     "ideal": [[4 * 256 * 32, 256 * 32]],
     "mistralai/Mistral-7B-v0.1/TP1": [

benchmarks/kernels/graph_machete_bench.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import math
 import pickle
 import re

benchmarks/kernels/utils.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import dataclasses
 from typing import Any, Callable, Iterable, Optional
 

benchmarks/kernels/weight_shapes.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 # Weight Shapes are in the format
 # ([K, N], TP_SPLIT_DIM)
 # Example:

benchmarks/overheads/benchmark_hashing.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import cProfile
 import pstats
 

cmake/hipify.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 #!/usr/bin/env python3
 
 #

collect_env.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 # ruff: noqa
 # code borrowed from https://github.com/pytorch/pytorch/blob/main/torch/utils/collect_env.py
 

csrc/cutlass_extensions/vllm_cutlass_library_extension.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import enum
 from typing import Dict, Union
 

csrc/quantization/machete/generate.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import itertools
 import math
 import os

docs/source/conf.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 # Configuration file for the Sphinx documentation builder.
 #
 # This file only contains a selection of the most common options. For a full

docs/source/generate_examples.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import itertools
 import re
 from dataclasses import dataclass, field

examples/offline_inference/aqlm_example.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 from vllm import LLM, SamplingParams
 from vllm.utils import FlexibleArgumentParser
 

examples/offline_inference/arctic.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 from vllm import LLM, SamplingParams
 
 # Sample prompts.

examples/offline_inference/audio_language.py

@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: Apache-2.0
 """
 This example shows how to use vLLM for running offline inference 
 with the correct prompt format on audio language models.

examples/offline_inference/basic.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 from vllm import LLM, SamplingParams
 
 # Sample prompts.

examples/offline_inference/basic_with_model_default_sampling.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 from vllm import LLM
 
 # Sample prompts.

examples/offline_inference/chat.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 from vllm import LLM, SamplingParams
 
 llm = LLM(model="meta-llama/Meta-Llama-3-8B-Instruct")

examples/offline_inference/chat_with_tools.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 # ruff: noqa
 import json
 import random

examples/offline_inference/classification.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 from vllm import LLM
 
 # Sample prompts.

examples/offline_inference/cli.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 from dataclasses import asdict
 
 from vllm import LLM, SamplingParams

examples/offline_inference/cpu_offload.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 from vllm import LLM, SamplingParams
 
 # Sample prompts.

examples/offline_inference/distributed.py

@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: Apache-2.0
 """
 This example shows how to use Ray Data for running offline batch inference
 distributively on a multi-nodes cluster.

examples/offline_inference/embedding.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 from vllm import LLM
 
 # Sample prompts.

examples/offline_inference/encoder_decoder.py

@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: Apache-2.0
 '''
 Demonstrate prompting of text-to-text
 encoder/decoder models, specifically BART

examples/offline_inference/florence2_inference.py

@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: Apache-2.0
 '''
 Demonstrate prompting of text-to-text
 encoder/decoder models, specifically Florence-2

examples/offline_inference/gguf_inference.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 from huggingface_hub import hf_hub_download
 
 from vllm import LLM, SamplingParams

examples/offline_inference/llm_engine_example.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import argparse
 from typing import List, Tuple
 

examples/offline_inference/lora_with_quantization_inference.py

@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: Apache-2.0
 """
 This example shows how to use LoRA with different quantization techniques
 for offline inference.

examples/offline_inference/mlpspeculator.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import gc
 import time
 from typing import List

examples/offline_inference/multilora_inference.py

@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: Apache-2.0
 """
 This example shows how to use the multi-LoRA functionality
 for offline inference.

examples/offline_inference/neuron.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 from vllm import LLM, SamplingParams
 
 # Sample prompts.

examples/offline_inference/neuron_int8_quantization.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import os
 
 from vllm import LLM, SamplingParams

examples/offline_inference/pixtral.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 # ruff: noqa
 import argparse
 

examples/offline_inference/prefix_caching.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 from vllm import LLM, SamplingParams
 from vllm.distributed import cleanup_dist_env_and_memory
 

examples/offline_inference/profiling.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import inspect
 import json
 import os

examples/offline_inference/profiling_tpu/profiling.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import argparse
 import dataclasses
 import os

examples/offline_inference/rlhf.py

@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: Apache-2.0
 """
 a simple demonstration of RLHF with vLLM, inspired by
 the OpenRLHF framework https://github.com/OpenRLHF/OpenRLHF .

examples/offline_inference/save_sharded_state.py

@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: Apache-2.0
 """
 Saves each worker's model state dict directly to a checkpoint, which enables a
 fast load path for large tensor-parallel models where each worker only needs to

examples/offline_inference/scoring.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 from vllm import LLM
 
 # Sample prompts.

examples/offline_inference/simple_profiling.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import os
 import time
 

examples/offline_inference/structured_outputs.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 from enum import Enum
 
 from pydantic import BaseModel

examples/offline_inference/torchrun_example.py

@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: Apache-2.0
 """
 experimental support for tensor-parallel inference with torchrun,
 see https://github.com/vllm-project/vllm/issues/11400 for

examples/offline_inference/tpu.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 from vllm import LLM, SamplingParams
 
 prompts = [

examples/offline_inference/vision_language.py

@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: Apache-2.0
 """
 This example shows how to use vLLM for running offline inference with
 the correct prompt format on vision language models for text generation.

examples/offline_inference/vision_language_embedding.py

@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: Apache-2.0
 """
 This example shows how to use vLLM for running offline inference with
 the correct prompt format on vision language models for multimodal embedding.

examples/offline_inference/vision_language_multi_image.py

@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: Apache-2.0
 """
 This example shows how to use vLLM for running offline inference with
 multi-image input on vision language models for text generation,

examples/offline_inference/whisper.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import time
 
 from vllm import LLM, SamplingParams

examples/online_serving/api_client.py

@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: Apache-2.0
 """Example Python client for `vllm.entrypoints.api_server`
 NOTE: The API server is used only for demonstration and simple performance
 benchmarks. It is not intended for production use.

examples/online_serving/cohere_rerank_client.py

@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: Apache-2.0
 """
 Example of using the OpenAI entrypoint's rerank API which is compatible with
 the Cohere SDK: https://github.com/cohere-ai/cohere-python

examples/online_serving/gradio_openai_chatbot_webserver.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import argparse
 
 import gradio as gr

examples/online_serving/gradio_webserver.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import argparse
 import json
 

examples/online_serving/jinaai_rerank_client.py

@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: Apache-2.0
 """
 Example of using the OpenAI entrypoint's rerank API which is compatible with
 Jina and Cohere https://jina.ai/reranker

examples/online_serving/openai_chat_completion_client.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 from openai import OpenAI
 
 # Modify OpenAI's API key and API base to use vLLM's API server.

examples/online_serving/openai_chat_completion_client_for_multimodal.py

@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: Apache-2.0
 """An example showing how to use vLLM to serve multimodal models 
 and run online serving with OpenAI client.
 

examples/online_serving/openai_chat_completion_client_with_tools.py

@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: Apache-2.0
 """
 Set up this example by starting a vLLM OpenAI-compatible server with tool call
 options enabled. For example:

examples/online_serving/openai_chat_completion_structured_outputs.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 from enum import Enum
 
 from openai import OpenAI

examples/online_serving/openai_chat_completion_with_reasoning.py

@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: Apache-2.0
 """
 An example shows how to generate chat completions from reasoning models
 like DeepSeekR1.

examples/online_serving/openai_chat_completion_with_reasoning_streaming.py

@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: Apache-2.0
 """
 An example shows how to generate chat completions from reasoning models
 like DeepSeekR1.

examples/online_serving/openai_chat_embedding_client_for_multimodal.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import argparse
 import base64
 import io

examples/online_serving/openai_completion_client.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 from openai import OpenAI
 
 # Modify OpenAI's API key and API base to use vLLM's API server.

examples/online_serving/openai_cross_encoder_score.py

@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: Apache-2.0
 """
 Example online usage of Score API.
 

examples/online_serving/openai_embedding_client.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 from openai import OpenAI
 
 # Modify OpenAI's API key and API base to use vLLM's API server.

examples/online_serving/openai_pooling_client.py

@@ -1,3 +1,4 @@
+# SPDX-License-Identifier: Apache-2.0
 """
 Example online usage of Pooling API.
 

examples/online_serving/opentelemetry/dummy_client.py

@@ -1,3 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+
 import requests
 from opentelemetry.exporter.otlp.proto.grpc.trace_exporter import (
     OTLPSpanExporter)