Source: https://seerepo.com/r/oobabooga/textgen/built
Markdown source: https://seerepo.com/r/oobabooga/textgen/built.md
Description: textgen is a Python repo built with FastAPI. It stores its data in chromadb. It has 29,113 written lines, with 1 test file.

# What is textgen built with?

textgen is a Python repo built with FastAPI. It stores its data in chromadb, uses AI through bitsandbytes, coqui-tts, einops, openai-whisper, sentencepiece, and 13 more, makes its content with lxml, odfpy, openpyxl, pillow, beautifulsoup4, and 5 more, validates its data with Pydantic, and uses Hugging Face. It has 29,113 written lines, with 1 test file.

Read on November 5, 2026, at commit c93f887.

Repository: https://github.com/oobabooga/textgen

## Frontend

### Interface

- **[Rich](https://seerepo.com/uses/rich)**: 1 file

## Backend

### The base

- **[FastAPI](https://seerepo.com/uses/fastapi)**: 1 file

### Data

- **[chromadb](https://seerepo.com/uses/chromadb)**: 2 files

### AI

- **bitsandbytes**
- **coqui-tts**
- **[einops](https://seerepo.com/uses/einops)**
- **openai-whisper**
- **[sentencepiece](https://seerepo.com/uses/sentencepiece)**
- **torchao**
- **[transformers](https://seerepo.com/uses/transformers)**: 13 files
- **[accelerate](https://seerepo.com/uses/accelerate)**: 2 files
- **[datasets](https://seerepo.com/uses/datasets)**: 2 files
- **[peft](https://seerepo.com/uses/peft)**: 2 files
- **[sentence-transformers](https://seerepo.com/uses/sentence-transformers)**: 2 files
- **diffusers**: 1 file
- **[mcp](https://seerepo.com/uses/mcp)**: 1 file
- **[nltk](https://seerepo.com/uses/nltk)**: 1 file
- **Optuna**: 1 file
- **[safetensors](https://seerepo.com/uses/safetensors)**: 1 file
- **spaCy**: 1 file
- **[tiktoken](https://seerepo.com/uses/tiktoken)**: 1 file

### Services

- **[Hugging Face](https://seerepo.com/uses/hugging-face)**: 2 files · [download-model.py, line 49](https://github.com/oobabooga/textgen/blob/c93f8871239550de2ccfe1e95d469aa82616f07e/download-model.py#L49)

## Support

### Content

- **[lxml](https://seerepo.com/uses/lxml)**
- **odfpy**
- **[openpyxl](https://seerepo.com/uses/openpyxl)**
- **[pillow](https://seerepo.com/uses/pillow)**: 7 files
- **[beautifulsoup4](https://seerepo.com/uses/beautifulsoup4)**: 2 files
- **[markdown](https://seerepo.com/uses/markdown)**: 2 files
- **[pymupdf](https://seerepo.com/uses/pymupdf)**: 2 files
- **[python-docx](https://seerepo.com/uses/python-docx)**: 2 files
- **[jinja2](https://seerepo.com/uses/jinja2)**: 1 file
- **[python-pptx](https://seerepo.com/uses/python-pptx)**: 1 file

### Validation

- **[Pydantic](https://seerepo.com/uses/pydantic)**: 2 files

### Tools

- **ipython**
- **tensorboard**

### Other

- **audioop-lts**
- **omegaconf**
- **SciPy**
- **soundfile**
- **wandb**
- **NumPy**: 9 files
- **pyyaml**: 7 files
- **requests**: 4 files
- **pandas**: 2 files
- **pydub**: 2 files
- **tqdm**: 2 files
- **ddgs**: 1 file
- **sse-starlette**: 1 file
- **trafilatura**: 1 file

## Other repos with a similar stack

- [bojieli/ai-agent-book](https://seerepo.com/r/bojieli/ai-agent-book/built): FastAPI, chromadb, Hugging Face, Rich, bitsandbytes, openai-whisper, sentencepiece, transformers, accelerate, datasets, peft, sentence-transformers, MCP, nltk, tiktoken, lxml, openpyxl, pillow, beautifulsoup4, markdown, pymupdf, python-docx, python-pptx, Pydantic · [Compare](https://seerepo.com/compare/bojieli/ai-agent-book/vs/oobabooga/textgen)
- [datawhalechina/hello-agents](https://seerepo.com/r/datawhalechina/hello-agents/built): FastAPI, chromadb, Hugging Face, Rich, transformers, accelerate, datasets, peft, sentence-transformers, MCP, safetensors, spaCy, tiktoken, lxml, openpyxl, pillow, beautifulsoup4, markdown, pymupdf, python-docx, jinja2, Pydantic · [Compare](https://seerepo.com/compare/datawhalechina/hello-agents/vs/oobabooga/textgen)
- [unslothai/unsloth](https://seerepo.com/r/unslothai/unsloth/built): FastAPI, Hugging Face, Rich, einops, openai-whisper, sentencepiece, transformers, accelerate, datasets, peft, sentence-transformers, diffusers, safetensors, tiktoken, lxml, pillow, pymupdf, python-docx, jinja2, Pydantic · [Compare](https://seerepo.com/compare/oobabooga/textgen/vs/unslothai/unsloth)
- [mudler/LocalAI](https://seerepo.com/r/mudler/LocalAI/built): FastAPI, Hugging Face, bitsandbytes, coqui-tts, einops, sentencepiece, torchao, transformers, accelerate, datasets, peft, sentence-transformers, diffusers, MCP, safetensors, tiktoken, pillow, jinja2 · [Compare](https://seerepo.com/compare/mudler/LocalAI/vs/oobabooga/textgen)
- [run-llama/llama_index](https://seerepo.com/r/run-llama/llama_index/built): chromadb, Rich, einops, sentencepiece, transformers, datasets, peft, sentence-transformers, MCP, nltk, tiktoken, lxml, pillow, beautifulsoup4, pymupdf, jinja2, Pydantic · [Compare](https://seerepo.com/compare/oobabooga/textgen/vs/run-llama/llama_index)
- [jianchang512/pyvideotrans](https://seerepo.com/r/jianchang512/pyvideotrans/built): Hugging Face, Rich, bitsandbytes, openai-whisper, sentencepiece, transformers, accelerate, datasets, peft, diffusers, safetensors, tiktoken, pillow, markdown, jinja2, Pydantic · [Compare](https://seerepo.com/compare/jianchang512/pyvideotrans/vs/oobabooga/textgen)

## Frequently asked questions

### What language is textgen written in?

textgen is written in Python, with 29,113 written lines.

### What framework does textgen use?

textgen is built with FastAPI.

### What database does textgen use?

textgen stores its data in chromadb.

### Does textgen use AI?

Yes. textgen uses AI through bitsandbytes, coqui-tts, einops, openai-whisper, sentencepiece, torchao, transformers, accelerate, datasets, peft, sentence-transformers, diffusers, mcp, nltk, Optuna, safetensors, spaCy, and tiktoken.

### What services does textgen use?

textgen uses Hugging Face.

The proof: [download-model.py, line 49](https://github.com/oobabooga/textgen/blob/c93f8871239550de2ccfe1e95d469aa82616f07e/download-model.py#L49)

### Does textgen have tests?

Yes. textgen has 1 test file.

### Which repos have a stack like textgen's?

The ones that share most of its stack: bojieli/ai-agent-book, datawhalechina/hello-agents, and unslothai/unsloth.

## 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.