Source: https://seerepo.com/compare/bojieli/ai-infra-book/vs/vllm-project/vllm
Markdown source: https://seerepo.com/compare/bojieli/ai-infra-book/vs/vllm-project/vllm.md
Description: ai-infra-book and vllm share 39 technologies: FastAPI, gitdb, pyzmq, Hugging Face, starlette, Ray, AIOHTTP, mkdocs-material, apache-tvm-ffi, cuda-python, cupy-cuda13x, flashinfer-python, MCP, mistral-common, nixl, nvidia-cudnn-frontend, OpenAI, sentencepiece, tiktoken, timm, torch, transformers, torchvision, triton, safetensors, einops, torchaudio, cuda-bindings, tokenizers, markupsafe, markdown, pillow, jinja2, annotated-types, msgspec, Pydantic, OpenTelemetry, psutil, and prometheus-client. Both are written in Python.

# ai-infra-book vs vllm

ai-infra-book and vllm share 39 technologies: FastAPI, gitdb, pyzmq, Hugging Face, starlette, Ray, AIOHTTP, mkdocs-material, apache-tvm-ffi, cuda-python, cupy-cuda13x, flashinfer-python, MCP, mistral-common, nixl, nvidia-cudnn-frontend, OpenAI, sentencepiece, tiktoken, timm, torch, transformers, torchvision, triton, safetensors, einops, torchaudio, cuda-bindings, tokenizers, markupsafe, markdown, pillow, jinja2, annotated-types, msgspec, Pydantic, OpenTelemetry, psutil, and prometheus-client. Both are written in Python.

- [bojieli/ai-infra-book](https://seerepo.com/r/bojieli/ai-infra-book): 《深入理解 AI Infra：量化分析与系统设计》（李博杰 著）开源书稿：从硬件约束和模型架构出发，量化推导 LLM 推理与训练系统设计。含全书正文、PDF、配套计算工具与实验
- [vllm-project/vllm](https://seerepo.com/r/vllm-project/vllm): A high-throughput and memory-efficient inference and serving engine for LLMs

## What they share and what sets them apart

- In common: FastAPI, gitdb, pyzmq, Hugging Face, starlette, Ray, AIOHTTP, mkdocs-material, apache-tvm-ffi, cuda-python, cupy-cuda13x, flashinfer-python, MCP, mistral-common, nixl, nvidia-cudnn-frontend, OpenAI, sentencepiece, tiktoken, timm, torch, transformers, torchvision, triton, safetensors, einops, torchaudio, cuda-bindings, tokenizers, markupsafe, markdown, pillow, jinja2, annotated-types, msgspec, Pydantic, OpenTelemetry, psutil, prometheus-client
- Only in ai-infra-book: blobfile, huey, pyarrow, Redis, s3fs, s3transfer, smart-open, tensorboard-data-server, transferqueue, Anthropic, anyio, Flask, flask-cors, graphene, graphql-core, graphql-relay, hypercorn, jupyter-core, kui, Lightning, quart, tornado, uvicorn, werkzeug, cycler, fonttools, blessed, contourpy, Gradio, kaleido, matplotlib-inline, Plotly, prompt-toolkit, pyecharts, seaborn, matplotlib, alembic, SQLAlchemy, cryptography, itsdangerous, Google, accelerate, comet-ml, cuda-core, cuda-toolkit, decord2, descript-audio-codec, e2b, einx, emerging-optimizers, evaluate, fla-core, flash-linear-attention, flashinfer-cubin, jaxlib, llvmlite, loralib, mamba-ssm, math-verify, megatron-bridge, ml-dtypes, MLflow, mlx, mlx-lm, mooncake-transfer-engine-cuda13, nltk, nvidia-cublas, nvidia-cudnn-cu12, nvidia-cudnn-cu13, nvidia-cufft, nvidia-cusolver, nvidia-cusparse, nvshmem4py-cu13, onnx-ir, onnxscript, open-clip-torch, openai-agents, opt-einsum, peft, pythainlp, qwen-vl-utils, rouge-score, sacrebleu, scikit-learn, sglang-router, silero-vad, sklearn-crfsuite, skops, spaCy, spacy-legacy, stanza, thinc, torch-c-dlpack-ext, tensordict, flash-attn, megatron-core, datasets, xformers, ONNX, deepspeed, diffusers, jax, librosa, causal-conv1d, ModelScope, nvidia-cuda-runtime, torchdata, transformer-engine, cairosvg, defusedxml, imageio, imageio-ffmpeg, lxml, mako, markdown-it-py, mdurl, nbformat, pylatexenc, pytablewriter, python-magic, pymupdf, av, dacite, fastjsonschema, jsonschema, jsonschema-specifications, pathvalidate, referencing, typepy, aiobotocore, boto3, botocore, colorlog, docker, gpustat, Gunicorn, opencensus-context, opentelemetry-exporter-prometheus, opentelemetry-semantic-conventions, spacy-loggers, swanlab, loguru
- Only in vllm: axum, Tokio, tower, indicatif, AMD, auto-round-lib, conch-triton-kernels, cupy-cuda12x, humming-kernels, lm-format-enforcer, mooncake-transfer-engine, mooncake-transfer-engine-rocm, NVIDIA, openai-harmony, oss-harmony, quack-kernels, vllm-bench, vllm-chat, vllm-engine-core-client, vllm-llm, vllm-managed-engine, vllm-parser, vllm-proto, vllm-server, vllm-text, vllm-tokenizer, xgrammar-structural-tag, llm-multimodal, compressed-tensors, xgrammar, partial-json-parser, lmcache, vllm-xpu-kernels, runai-model-streamer, tokenspeed-mla, llguidance, outlines-core, tpu-inference, fastokens, fastsafetensors, hf-hub, instanttensor, riptoken, tekken, tensorizer, tiktoken-rs, mkdocs-material-extensions, parquet, properdocs, pymdown-extensions, selectolax, xlsxwriter, image, minijinja, torchcodec, minijinja-contrib, validator, opentelemetry-semantic-conventions-ai, python-json-logger, vllm-metrics, vllm-tracing, tracing, tracing-futures, tracing-subscriber, model-hosting-container-standards, prometheus-fastapi-instrumentator, tracing-log

## Side by side

| | ai-infra-book | vllm |
|---|---|---|
| Size | 1,438,355 lines written, 1,748 code files | 2,606,460 lines written, 3,630 code files |
| Who writes it | 16.6% of its commits signed by AI. With Claude Code and Codex | 14.7% of its commits signed by AI. With Claude Code, Codex, Gemini CLI, Cursor, GitHub Copilot, Devin, Mistral Vibe, CodeRabbit, Factory, and Qwen Code |
| Activity | Last push October 1, 2026, 314 commits, 8,032 stars | Last push November 17, 2026, 25,558 commits, 97,060 stars |
| Tests | 1,093 test files. 17.8% of its code files have a test that imports them | 2,586 test files. 60.7% of its code files have a test that imports them |

- Size: Lines written by people and agents, without generated ones; code files, without tests.
- Who writes it: A commit counts as AI's when it carries its signature: a Co-Authored-By line, its author name or the merge of its branch.
- Activity: From GitHub and its history, on the day SeeRepo read it.
- Tests: The files that test the code, and the share of code files a test imports.

## Frequently asked questions

### Which is bigger, ai-infra-book or vllm?

vllm, with 2,606,460 lines written; ai-infra-book has 1,438,355.

The proof: [vllm-project/vllm's code at commit 61c3895](https://github.com/vllm-project/vllm/tree/61c3895fc937b4ffc2a59abe188959de1e459c8a)

### Which has more GitHub stars, ai-infra-book or vllm?

vllm, with 97,060 stars; ai-infra-book has 8,032.

The proof: [vllm-project/vllm on GitHub](https://github.com/vllm-project/vllm)

### Which has more tests, ai-infra-book or vllm?

vllm, with 2,586 test files; ai-infra-book has 1,093.

The proof: [vllm-project/vllm's code at commit 61c3895](https://github.com/vllm-project/vllm/tree/61c3895fc937b4ffc2a59abe188959de1e459c8a)

### Which was updated more recently, ai-infra-book or vllm?

vllm: its last push was on November 17, 2026, and ai-infra-book's on October 1, 2026.

The proof: [vllm-project/vllm's commits on GitHub](https://github.com/vllm-project/vllm/commits)

### Which has more commits signed by AI, ai-infra-book or vllm?

ai-infra-book: an AI signed 16.6% of its commits, with Claude Code and Codex; in vllm, 14.7%.

The proof: [the commits each tool signed in bojieli/ai-infra-book](https://seerepo.com/r/bojieli/ai-infra-book/ai)

## Sources

- bojieli/ai-infra-book: Read on November 17, 2026, at commit 3bdcb4f (https://github.com/bojieli/ai-infra-book/tree/3bdcb4fcab73010eeaccf31cc10a8242a895d82e)
- vllm-project/vllm: Read on November 17, 2026, at commit 61c3895 (https://github.com/vllm-project/vllm/tree/61c3895fc937b4ffc2a59abe188959de1e459c8a)

## What is your repo built with?

Any public repo, with no account: https://seerepo.com/r/owner/repo/built, with the owner and name it has on GitHub.
