Granite 3 models

9 Granite 3 models from IBM on Hugging Face, from 1.3B to 8.2B parameters, published by IBM in BF16 and F32, with contexts from 4K to 128K tokens. The smallest official model, granite-3.0-1b-a400m-instruct, needs about 3.0 GB of VRAM at its published precision; the cheapest live fit is RTX 3070 at $0.088/hr.

Part of the Granite series · Next generation: Granite 4

Pick a size

One row per official Granite 3 size: the VRAM it needs at each precision, the cheapest GPU that holds it today, and the KV cache for a 32K-token context.

ModelParametersNative VRAMFP8 VRAMINT4 VRAMLive GPU fit (native)Est. $/hrKV cache at 32K
granite-3.0-1b-a400m-instruct1.3B3.0 GB1.5 GB0.7 GBRTX 3070$0.088/hr1.5 GB
granite-3.0-1b-a400m-base1.4B6.2 GB1.5 GB0.8 GBRTX 3070$0.088/hr1.5 GB
granite-3.3-2b-instruct2.5B5.7 GB2.8 GB1.4 GBRTX 3070$0.088/hr2.5 GB
granite-guardian-3.3-8b8.2B18.3 GB9.1 GB4.6 GBRTX A5000$0.176/hr5.0 GB

VRAM is weights times a flat 1.2 overhead; the KV cache is a separate per-model figure at 16-bit, shown where the architecture is published. See the methodology.

Official models (9)

ModelParametersPublished asVRAM neededLive GPU fitEst. $/hr
granite-3.0-1b-a400m-instruct1.3BBF163.0 GBRTX 3070$0.088/hr
granite-3.0-1b-a400m-base1.4BF326.2 GBRTX 3070$0.088/hr
granite-3.1-2b-instruct2.5BBF165.7 GBRTX 3070$0.088/hr
granite-3.3-2b-instruct2.5BBF165.7 GBRTX 3070$0.088/hr
granite-3.0-8b-instruct8.2BBF1618.3 GBRTX A5000$0.176/hr
granite-3.1-8b-instruct8.2BBF1618.3 GBRTX A5000$0.176/hr
granite-3.2-8b-instruct8.2BBF1618.3 GBRTX A5000$0.176/hr
granite-3.3-8b-instruct8.2BBF1618.3 GBRTX A5000$0.176/hr
granite-guardian-3.3-8b8.2BBF1618.3 GBRTX A5000$0.176/hr

VRAM is for the precision the model is published in: the weights times a flat 1.2 overhead, with the KV cache not included. See the methodology. The fit shown is the lowest-priced single GPU type that holds the model at that precision, or the lowest-priced multi-GPU set (up to 8) when none does.

Run and fine-tune Granite 3

More from IBM

Start ComfyUI on a rented GPU.

© 2026 Aquanode. All rights reserved.

All trademarks, logos and brand names are the property of their respective owners.