llama-cpp-python has 689K monthly downloads and 10K GitHub stars. Version 0.3.33 supports Python 3.8+, first published in 2023.
It is actively maintained under a permissive license, last released in Jul 2026.
It is built on top of diskcache, jinja2 and numpy. It is used by 279 tracked packages, including khoj, lilbee and hugpy.
Package Insights
((week_daily_avg - month_daily_avg) / month_daily_avg) * 100Weekly Downloads
GitHub Stars
Downloads by OS
Python Versions
Top Countries
Dependencies
- diskcache >=5.6.1
- jinja2 >=2.11.3
- numpy >=1.20.0
- typing-extensions >=4.5.0
22 optional dependencies
- fastapi[test]
- fastapi[server]
- httpx[dev]
- httpx[test]
- huggingface-hub[test]
- llama-cpp-python[all]
- mkdocs[dev]
- mkdocs-material[dev]
- mkdocstrings[dev]
- pydantic-settings[test]
- pydantic-settings[server]
- pytest[test]
- pytest[dev]
- pyyaml[server]
- ruff[dev]
- scipy[test]
- sse-starlette[test]
- sse-starlette[server]
- starlette-context[test]
- starlette-context[server]
- twine[dev]
- uvicorn[server]
Used By
- khoj
- lilbee
- hugpy
- sapiens-transformers
- utilities-nlp
- llama-cpp-agent
- llama-index-llms-llama-cpp
- instructlab
- maru-deep-pro-search
- logdetective
- outetts
- rbtr
- freegenius
- lore-ai
- programasweights
- mahavishnu
- transai
- bikescout
- locallab
- llama-cpp-haystack
- augllm
- modelpulse
- steadytext
- developergpt
- py-llm-core
- vikas-autocomplete
- ovos-gguf-embeddings-plugin
- cosma-backend
- dashai-deepseek-model-package
- local-llm
- langchain-llamacpp-chat-model
- vision-arwaky
- cortex-llm
- ledgermind
- sinapsis-llama-cpp
- gitai-local
- sapiens-dataset
- ownscribe
- allmdev
- cat-agent
- vital-llm-reasoner
- ovos-gguf-plugin
- swingft-test-v3
- vail-model-registry
- aaria-lm
- fusellm
- corp-extractor
- armor-ai
- dora-qwen
- nnzip