_____ _____ _ _ __ __ ______ _____ _ _
/ ____| __ \| | | | | \/ | ____|/ ____| | | |
| | __| |__) | | | | | \ / | |__ | (___ | |__| |
| | |_ | ___/| | | | | |\/| | __| \___ \| __ |
| |__| | | | |__| | | | | | |____ ____) | | | |
\_____|_| \____/ |_| |_|______|_____/|_| |_|
Free LLM inference
Use any OpenAI-compatible tool with community GPUs. No credit card, no limits.
8
Models online
1
Donors online
14
Requests today
How it works
1
Share
Donors run the agent, Ollama serves models. Anyone can contribute a GPU.
2
Match
The coordinator routes your request to an available donor with the model you need.
3
Use
Your tools just work — OpenAI-compatible API. No config changes needed.
Top models right now
qwen3.5:9b
live
qwen3.6:27b
live
GPT-OSS:20B-32k
live
VladimirGav/Qwen3.6-27B-16GB-VRAM-Uncensored:latest
live
gemma4:26b
live