What is ai-infra-book built with?
ai-infra-book is a Python repo built with FastAPI, anyio, Flask, flask-cors, graphene, and 13 more. It reaches its data with alembic and SQLAlchemy, signs people in with cryptography, itsdangerous, and Google, uses AI through accelerate, apache-tvm-ffi, comet-ml, cuda-core, cuda-python, and 84 more, makes its content with cairosvg, defusedxml, imageio, imageio-ffmpeg, lxml, and 13 more, validates its data with annotated-types, dacite, fastjsonschema, jsonschema, jsonschema-specifications, and 5 more, is tested with responses, playwright, Pytest, and Rich, runs in production with aiobotocore, boto3, botocore, colorlog, docker, and 11 more, and uses blobfile, gitdb, huey, pyarrow, redis, and 7 more. It has 1,438,355 written lines, with 1,093 test files.
Read on November 17, 2026, at commit 3bdcb4f. Read it as markdown
The card says what it's built with. Its whole map opens it up: every part of the code and how it connects to the rest.
Frontend
Styles 3
Interface 10
matplotlib249 files
contourpy
Gradio
- Plotly
prompt-toolkit
pyecharts
seaborn
- blessed
- kaleido
- matplotlib-inline
Backend
The base 1
FastAPI3 files
Data 11
pyarrow
- redis
tensorboard-data-server
- pyzmq9 files
- blobfile
- gitdb
- huey
- s3fs
- s3transfer
- smart-open
- transferqueue
Data access 2
Sign-in 3
AI 88
- torch326 files
torchvision20 files
tensordict11 files
safetensors10 files
- ONNX4 files
deepspeed3 files
diffusers3 files
jax3 files
librosa3 files
torchaudio3 files
- ModelScope1 file
tokenizers1 file
accelerate
apache-tvm-ffi
cupy-cuda13x
descript-audio-codec
e2b
evaluate
fla-core
flash-linear-attention
flashinfer-cubin
flashinfer-python
jaxlib
mcp
- MLflow
mlx
mooncake-transfer-engine-cuda13
nltk
- nvidia-cublas
nvidia-cudnn-frontend
onnx-ir
onnxscript
openai
openai-agents
peft
pythainlp
rouge-score
- scikit-learn
skops
- spaCy
spacy-legacy
stanza
thinc
timm
- transformers51 files
- triton11 files
- einops9 files
- flash-attn9 files
- megatron-core8 files
- datasets5 files
- xformers5 files
- causal-conv1d1 file
- cuda-bindings1 file
- nvidia-cuda-runtime1 file
- torchdata1 file
- transformer-engine1 file
- comet-ml
- cuda-core
- cuda-python
- cuda-toolkit
- decord2
- einx
- emerging-optimizers
- llvmlite
- loralib
- mamba-ssm
- math-verify
- megatron-bridge
- mistral-common
- ml-dtypes
- mlx-lm
- nixl
- nvidia-cudnn-cu12
- nvidia-cudnn-cu13
- nvidia-cufft
- nvidia-cusolver
- nvidia-cusparse
- nvshmem4py-cu13
- open-clip-torch
- opt-einsum
- qwen-vl-utils
- sacrebleu
- sentencepiece
- sglang-router
- silero-vad
- sklearn-crfsuite
- tiktoken
- torch-c-dlpack-ext
Other repos with a similar stack
Ordered by how much of its stack they share with ai-infra-book, then by stars.
- run-llamallama_index52.9k
LlamaIndex is the document processing platform for AI
51 of 415 in commonCompare →
Frequently asked questions
What language is ai-infra-book written in?
What framework does ai-infra-book use?
What does ai-infra-book sign people in with?
Does ai-infra-book use AI?
What services does ai-infra-book use?
Does ai-infra-book have tests?
Which repos have a stack like ai-infra-book's?
See inside your code, today.
Paste your repo and see it drawn, part by part: a medium one in under a minute.
Free, no account.Private repo? Sign in with GitHub
Is it your repo? Add its badge, keep it up to date or take it down →

