Model Intelligence Sheet
ibm-granite/granite-4.2-3b-GGUF overview
Granite 4.2 3B GGUF NOTE This repository contains the GGUF version of Granite 4.2 3B . Please refer to the original model's model card for additional details: …
Runs locally from ~1.36 GB disk (4 GB VRAM class GPUs with llama.cpp / guIDE).
Repository Files & Downloads
15 GGUF files detected
Direct downloads for local inference
| File | Type | Quantization | Size | Link |
|---|---|---|---|---|
| granite-4.2-3b-Q2_K.gguf | GGUF | Q2_K | 1.36 GB | Download |
| granite-4.2-3b-Q3_K_L.gguf | GGUF | Q3_K_L | 1.84 GB | Download |
| granite-4.2-3b-Q3_K_M.gguf | GGUF | Q3_K_M | 1.71 GB | Download |
| granite-4.2-3b-Q3_K_S.gguf | GGUF | Q3_K_S | 1.56 GB | Download |
| granite-4.2-3b-Q4_0.gguf | GGUF | Q4_0 | 1.98 GB | Download |
| granite-4.2-3b-Q4_1.gguf | GGUF | Q4_1 | 2.18 GB | Download |
| granite-4.2-3b-Q4_K_M.gguf | GGUF | Q4_K_M | 2.09 GB | Download |
| granite-4.2-3b-Q4_K_S.gguf | GGUF | Q4_K_S | 2.00 GB | Download |
| granite-4.2-3b-Q5_0.gguf | GGUF | Q5_0 | 2.38 GB | Download |
| granite-4.2-3b-Q5_1.gguf | GGUF | Q5_1 | 2.58 GB | Download |
| granite-4.2-3b-Q5_K_M.gguf | GGUF | Q5_K_M | 2.43 GB | Download |
| granite-4.2-3b-Q5_K_S.gguf | GGUF | Q5_K_S | 2.38 GB | Download |
| granite-4.2-3b-Q6_K.gguf | GGUF | Q6_K | 2.80 GB | Download |
| granite-4.2-3b-Q8_0.gguf | GGUF | Q8_0 | 3.63 GB | Download |
| granite-4.2-3b-bf16.gguf | GGUF | BF16 | 6.82 GB | Download |
Model Details
| Model ID | ibm-granite/granite-4.2-3b-GGUF |
|---|---|
| Author | ibm-granite |
| Pipeline | — |
| License | apache-2.0 |
| Base model | ibm-granite/granite-4.2-3b |
| Last modified | 2026-09-02T22:06:48.000Z |
Model README
---
license: apache-2.0
library_name: transformers
tags:
- language
- granite-4.2
- gguf
base_model:
- ibm-granite/granite-4.2-3b
---
Granite-4.2-3B (GGUF)
> [!NOTE]
> This repository contains the GGUF version of Granite-4.2-3B.
>
> Please refer to the original model's model card for additional details:
> https://huggingface.co/ibm-granite/granite-4.2-3b
Run ibm-granite/granite-4.2-3b-GGUF with guIDE
Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.
Source: Hugging Face · Compare models