Models / granite-3.0

granite-3.0 by ibm-granite

2 models (1B to 8B) · ↓ 95.4K downloads a month · from $0.10/h on RTX 3060 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.

granite-3.0 1B

Version	Runtime	Weights	GPU memory	Cheapest machine now
Originalibm-granite/granite-3.0-1b-a400m-base	vLLM	2.8 GB	5.5 GB	$0.10/h RTX 3060
instructibm-granite/granite-3.0-1b-a400m-instruct	vLLM	2.7 GB	5.4 GB	$0.10/h RTX 3060

granite-3.0 8B

Version	Runtime	Weights	GPU memory	Cheapest machine now
Originalibm-granite/granite-3.0-8b-base	vLLM	16 GB	21.1 GB	$0.18/h 2x RTX 3060
instructibm-granite/granite-3.0-8b-instruct	vLLM	16 GB	21.1 GB	$0.18/h 2x RTX 3060

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…