Technologies
What is verl built with?
verl is a Python repo built with FastAPI and uvicorn. It uses AI through transformers, accelerate, datasets, and einops, makes its content with av and pylatexenc, is tested with pytest-asyncio, pytest-rerunfailures, and Pytest, and uses pyarrow. It has 359 files and 187,062 written lines, with 304 test files.
Read on November 4, 2026, at commit 7454326. Read it as markdown
Backend
The base
- FastAPI2 files
Data
pyarrow
AI
- transformers36 files
- accelerate3 files
- datasets1 file
- einops1 file
Support
The base
uvicorn2 files
Content
- av
- pylatexenc1 file
Testing
pytest-asyncio
- pytest-rerunfailures
- Pytest172 files
Tools
- pre-commit
- tensorboard
- packaging19 files
Other
- hydra-core
- NumPy35 files
- pandas5 files
Other repos with a similar stack
Ordered by how much of its stack they share with verl, then by stars.
- open-webuiopen-webui157.7k
User-friendly AI Interface (Supports Ollama, OpenAI API, ...)
3 of 18 in common · SvelteCompare → - harry0703MoneyPrinterTurbo138.5k
利用 AI 大模型和自动化工作流,根据主题或关键词一键生成高清短视频。Generate HD short videos from a topic or keyword with an automated AI workflow.
3 of 18 in commonCompare →
Frequently asked questions
What language is verl written in?
What framework does verl use?
Does verl use AI?
What services does verl use?
Does verl have tests?
Which repos have a stack like verl's?
What is your repo built with?
Paste your public repo and see it drawn.
Free, no account.
Is it your repo? Add its badge, keep it up to date or take it down →