Granite-4.2-3B-BE-GGUF
Name Quant Method Bits Size Use Case --------------- Granite-4.2-3B-BE-Q2K.gguf Q2K 2 1.4G smallest K-quant, significant quality loss Granite-4.2-3B-BE-Q3KL.gguf Q3KL 3 1.8G very small, high quality loss Granite-4.2-3B-BE-Q3KM.gguf Q3KM 3 1.7G very small, high quality loss Granite-4.2-3B-BE-Q3KS.gguf Q3KS 3 1.6G very small, high quality loss Granite-4.2-3B-BE-IQ4NL.gguf IQ4NL 4 2.0G small, good quality for its size G...
Params
3.0 B
Context
—
Downloads 30d
144 K
Likes
1
Download history
tracking started — chart appears after 7 days of snapshots (1 recorded)144 K downloads in the last 30 days
Can you run it?
Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.
| File | Quant | Size | Est. VRAM | Verdict on RTX 4090 · 24 GB |
|---|---|---|---|---|
| Granite-4.2-3B-BE-Q2_K.gguf | Q2_K | 1.5 GB | 2.6 GB | ✅ Runs comfortably |
| Granite-4.2-3B-BE-Q3_K_S.gguf | Q3_K_S | 1.7 GB | 2.8 GB | ✅ Runs comfortably |
| Granite-4.2-3B-BE-Q3_K_M.gguf | Q3_K_M | 1.8 GB | 3.0 GB | ✅ Runs comfortably |
| Granite-4.2-3B-BE-Q3_K_L.gguf | Q3_K_L | 2.0 GB | 3.1 GB | ✅ Runs comfortably |
| Granite-4.2-3B-BE-IQ4_XS.gguf | IQ4_XS | 2.0 GB | 3.2 GB | ✅ Runs comfortably |
| Granite-4.2-3B-BE-IQ4_NL.gguf | IQ4_NL | 2.1 GB | 3.3 GB | ✅ Runs comfortably |
| Granite-4.2-3B-BE-Q4_0.gguf | Q4_0 | 2.1 GB | 3.3 GB | ✅ Runs comfortably |
| Granite-4.2-3B-BE-F32.gguf | GGUF | 14.6 GB | 17.1 GB | ✅ Runs comfortably |
Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.
Run it
copy-paste, exact tags checked against the Hub$ ollama run granite-4-2-3b-be-gguf # pin the quantization explicitly $ ollama run granite-4-2-3b-be-gguf-q4_0
$ huggingface-cli download taronaeo/Granite-4.2-3B-BE-GGUF-GGUF \
Granite-4.2-3B-BE-Q4_0.gguf --local-dir .
$ llama-cli -m Granite-4.2-3B-BE-Q4_0.gguf \
-c 8192 -ngl 99 -t 8 --color
$ curl -s https://aimodelscomparison.com/api/v1/models/granite-4-2-3b-be-gguf
{
"hf_id": "taronaeo/Granite-4.2-3B-BE-GGUF",
"params_b": 3.00,
"context_length": null,
"license": { "id": "apache-2.0", "commercial": "yes" },
"downloads_30d": 143923,
"vram_estimates": [
{ "quant": "GGUF", "gb": 17.1 },
{ "quant": "IQ4_NL", "gb": 3.3 }
],
"updated_at": "2026-11-10T01:00:40Z"
}
Specifications
- Parameters
- 3.0 B
- Licence
- apache-2.0
- First seen on the Hub
- 2026-10-18
- Training datasets
- undisclosed
- Added to our catalog
- 2026-11-10
Compare with any text-generation model