ahtx/kindi
Runs Bonsai language models on low-cost GPUs for inference.
- Stars
- 113
- Forks
- 29
- Open issues
- 12
- Last push
- 1d ago
kindi star history
+53 this week
Get kindi running
Setup: Easy · install script, typically about 5 min
curl -fsSL https://amirhusain.com/kindi/install.sh | bashCan I sell what I build with it?
Yes, with conditions
A guide, not legal advice.
About kindi
kindi is a free, open-source AI inference tool by ahtx, written mainly in Python. Runs Bonsai language models on low-cost GPUs for inference. It has 113 GitHub stars and 29 forks, and was last updated 1d ago. It's released under the AGPL-3.0 license, a standard open-source license. According to the project, it needs a GPU with at least 1 GB of memory.
- License
- AGPL-3.0
- Commercial use
- Copyleft
- Language
- Python
- Stars
- 113
- Last update
- 1d ago
- Runs on Mac
- Not stated
- GPU
- 1 GB+ (as stated by the project)
- Paid API key
- Not required
- Self-hostable
- Yes
Similar free tools
vllm96k stars · High-throughput inference engine for LLMs.
llm-action25k stars · 本项目旨在分享大模型相关技术原理以及实战经验（大模型工程化、大模型应用落地）
mlc-llm23k stars · Compile and deploy large language models on various platforms.
airllm42k stars · Runs large language models on low VRAM GPUs.
TensorRT-LLM15k stars · Optimizes inference for large language models on NVIDIA GPUs.
BitNet41k stars · Optimized inference framework for 1-bit LLMs on CPU and GPU.
kindi FAQ
Can I use kindi commercially?
With conditions. AGPL-3.0 allows commercial use, but if you modify it or build on it, even as a web service, your code must also be released under AGPL. This is a plain-English summary, not legal advice.
Is kindi free?
Yes. kindi is open source and free to use. It's released under the AGPL-3.0 license, a standard open-source license.
Does kindi run on a Mac?
The project doesn't say. Check its installation guide before you start.
Does kindi need a GPU?
According to the project, kindi needs a GPU with at least 1 GB of memory.
Is kindi still maintained?
Yes. It was last updated 1d ago and has 12 open issues on GitHub.
How hard is kindi to set up?
Easy: typically about 5 min, using install script. The project's install command is: curl -fsSL https://amirhusain.com/kindi/install.sh | bash
What are alternatives to kindi?
Similar free, open-source tools include vllm, llm-action, mlc-llm, airllm.
Maintain kindi?
Add this badge to your README to show it's featured on FreeAIRepo.
[](https://www.freeairepo.com/repo/ahtx/kindi)