GR

Granite 4.0 H 350M BE

GraniteGGUF by taronaeoUpdated 2026-10-18
Download on Hugging Face
Parameters340M
Context—
Layers—
Architecture—
Smallest quant0.2 GB
Quick answer

Granite 4.0 H 350M BE needs about 0.7 GB of memory at Q4_K_M (a 0.2 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 19 quants, from 0.2 GB (Q2_K) to 1.4 GB (F32).

Granite 4.0 H 350M BE – lightweight instruct model for on‑device use

Granite 4.0 H 350M BE is a 350 M parameter instruct model from IBM’s Granite family. It is finetuned with open source instruction data and reinforcement learning. The model supports multiple languages and is released under Apache 2.0.

Good at

  • Instruction following
  • Multilingual dialogue
  • Tool‑calling

Limits

  • No vision support
  • Requires at least 0.7 GB VRAM for Q2_K at 8k context
  • License is Apache 2.0

Best for

  • On‑device chat assistants
  • Fine‑tuning for niche domains
  • Research on lightweight LLMs

Who it’s for. Developers and researchers needing a compact, multilingual instruction model that can run on modest hardware. The model is suitable for edge deployments and rapid prototyping.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K0.2 GB 0.7 GB
Fits Get file →
Q3_K_S0.2 GB 0.7 GB
Fits Get file →
Q3_K_M0.2 GB 0.7 GB
Fits Get file →
Q3_K_L0.2 GB 0.7 GB
Fits Get file →
IQ4_XS0.2 GB 0.7 GB
Fits Get file →
Q4_00.2 GB 0.7 GB
Fits Get file →
IQ4_NL0.2 GB 0.7 GB
Fits Get file →
Q4_K_S0.2 GB 0.7 GB
Fits Get file →
Q4_K_M0.2 GB 0.7 GB
Fits Get file →
Q4_10.2 GB 0.8 GB
Fits Get file →
Q5_00.2 GB 0.8 GB
Fits Get file →
Q5_K_S0.2 GB 0.8 GB
Fits Get file →
Q5_K_M0.3 GB 0.8 GB
Fits Get file →
Q5_10.3 GB 0.8 GB
Fits Get file →
Q6_K0.3 GB 0.8 GB
Fits Get file →
Q8_00.4 GB 0.9 GB
Fits Get file →
F160.7 GB 1.3 GB
Fits Get file →
BF160.7 GB 1.3 GB
Fits Get file →
F321.4 GB 2.0 GB
Fits Get file →

For 8 GB VRAM use Q4_K_M (0.22 GB file, ~0.7 GB VRAM). If 4 GB VRAM is available, Q2_K (0.16 GB file, ~0.7 GB VRAM) is recommended. For higher quality, Q5_0 (0.25 GB file, ~0.8 GB VRAM) can be used on 8 GB VRAM systems.

What GPU do you need for Granite 4.0 H 350M BE?

These figures add the file size, an fp16 KV cache at an 8K context window and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: the best quant that fits is F32 (1.4 GB file, about 2.0 GB in use).

Which Granite 4.0 H 350M BE quant should you choose?

Q4_K_M is the usual starting point for Granite 4.0 H 350M BE: it balances file size and quality well. Here it is a 0.2 GB file.

The Q8_0 file is 0.4 GB, about 1.6 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

If memory is tight, the smallest file is Q2_K at 0.2 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Parameters340M

Source

Quantized bytaronaeo
FamilyGranite
Licenseapache-2.0
Downloads (HF)17,493
Last updated2026-10-18

Frequently asked questions about Granite 4.0 H 350M BE

How much VRAM does Granite 4.0 H 350M BE need?

About 0.7 GB at Q4_K_M with an 8K context window. That is the 0.2 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run Granite 4.0 H 350M BE on an 8 GB GPU?

Yes, with F32: the 1.4 GB file needs about 2.0 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Granite 4.0 H 350M BE GGUF?

It depends on the quant: Q2_K is 0.2 GB, Q3_K_S is 0.2 GB, Q3_K_M is 0.2 GB, Q3_K_L is 0.2 GB, IQ4_XS is 0.2 GB, Q4_0 is 0.2 GB and 13 more.

Where can I download Granite 4.0 H 350M BE GGUF?

The files are on Hugging Face in the taronaeo/Granite-4.0-H-350M-BE-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does Granite 4.0 H 350M BE use?

Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run Granite 4.0 H 350M BE on 4 GB VRAM?

Yes, the Q2_K quantization uses about 0.7 GB VRAM at 8k context, which fits within 4 GB VRAM limits.

Does the model support image inputs?

No, Granite 4.0 H 350M BE does not include vision capabilities.

What license applies to this model?

The model is released under the Apache 2.0 license.

Other GGUFs of granite-4.0-h-350m

Similar models

Browse all →

More from taronaeo

All by taronaeo →

More Granite models

All Granite →

Other tiny models

All tiny models →