ibm-granite
granite 4.2 3b
granite 4.2 3b is a 3.66B dense granite model from ibm-granite. Configured context length is 131,072 tokens. License: apache-2.0. Quant sizes are the GGUF file bytes published on Hugging Face when a file exists.
Parameters3.7B
Context128K
Attentiongqa
Licenseapache-2.0
40 layers · 8 KV heads · head dim 64 · GGUF repo · config
Quant file sizes
| Quant | File size | Source |
|---|---|---|
| BF16 | 6.82 GB | Hugging Face file bytes |
| Q8_0 | 3.63 GB | Hugging Face file bytes |
| Q6_K | 2.80 GB | Hugging Face file bytes |
| Q5_K_M | 2.43 GB | Hugging Face file bytes |
| Q5_K_S | 2.38 GB | Hugging Face file bytes |
| Q5_1 | 2.58 GB | Hugging Face file bytes |
| Q5_0 | 2.38 GB | Hugging Face file bytes |
| Q4_K_M | 2.09 GB | Hugging Face file bytes |
| Q4_K_S | 2.00 GB | Hugging Face file bytes |
| Q4_1 | 2.18 GB | Hugging Face file bytes |
| Q4_0 | 1.98 GB | Hugging Face file bytes |
| Q3_K_L | 1.84 GB | Hugging Face file bytes |
| Q3_K_M | 1.71 GB | Hugging Face file bytes |
| Q3_K_S | 1.56 GB | Hugging Face file bytes |
| Q2_K | 1.36 GB | Hugging Face file bytes |
On common machines
| Device | Verdict | Quant | Speed |
|---|---|---|---|
| GeForce RTX 4060 Ti 8GB | Runs | Q8_0 | 35–47 |
| GeForce RTX 3060 12GB | Runs | BF16 | 25–34 |
| GeForce RTX 4060 Ti 16GB | Runs great | BF16 | 20–27 |
| GeForce RTX 4090 | Runs great | BF16 | 69–95 |
| GeForce RTX 5090 | Runs great | BF16 | 123–168 |
| Apple M4 Pro | Runs great | BF16 | 19–26 |
| Apple M5 | Runs | BF16 | 11–14 |
| CPU only · DDR5-5600 | Runs on CPU | BF16 | 6.2–8.4 |
Smallest devices that fit it fully
- GeForce RTX 3050 6GB — Tight fit, Q8_0, 20–28 tok/s
- GeForce RTX 2060 6GB — Tight fit, Q8_0, 40–55 tok/s
- GeForce GTX 1660 Super — Tight fit, Q8_0, 40–55 tok/s
- GeForce GTX 1660 Ti — Tight fit, Q8_0, 35–47 tok/s
- Intel Arc A380 — Tight fit, Q8_0, 22–31 tok/s
- GeForce RTX 5060 Ti 8GB — Runs, Q8_0, 54–74 tok/s
Run it
Commands for the GeForce RTX 4060 Ti 16GB estimate (BF16).
llama.cpp
llama-server -hf ibm-granite/granite-4.2-3b-GGUF:BF16 -c 8192 -ngl 40
No matching name in the Ollama library snapshot.
LM Studio
lms get ibm-granite/granite-4.2-3b-GGUF # Then load quant BF16