Models / Jan-nano

Jan-nano by Menlo

GPU memory needed (smallest version of each): Jan-nano ~4.02B about 12.5 GB. The cheapest runs on RTX 4060 Ti from $0.12/h right now.

1 model (~4.02B) · ↓ 2K downloads a month · from $0.12/h on RTX 4060 Ti right now. Each model's versions do the same job; quantized ones (AWQ, FP8, 4-bit) need less GPU memory and run on cheaper machines.

Jan-nano ~4.02B

Version	Runtime	Weights	GPU memory	Cheapest machine now
OriginalMenlo/Jan-nano	vLLM	8.0 GB	12.5 GB	$0.12/h RTX 4060 Ti
128kMenlo/Jan-nano-128k	vLLM	8.0 GB	12.5 GB	$0.12/h RTX 4060 Ti

Model details come from Hugging Face and are refreshed daily; prices follow the live offers. GPU memory is GridPup's estimate: the weights, room for an 8,192-token conversation (chat models) and the runtime's own memory. Large models are split across the GPUs of one machine ("4x RTX 4090").

Loading…