granite-4.1-8b vs Llama-3.2-3B
Specs, VRAM requirements and download trends — updated 16 August 2026.
ibm-granite · A · Open weights
Params
8.8 B
Context
131,072
30d
192 K
apache-2.0 · commercial OK
meta-llama · B · Open weights
Params
3.2 B
Context
—
30d
269 K
llama3.2 licence · conditional
Specification comparison
Differences are highlighted; identical values are muted.
| Specification | granite-4.1-8b | Llama-3.2-3B |
|---|---|---|
| Parameters | 8.8 B | 3.2 B |
| Architecture | GraniteForCausalLM | LlamaForCausalLM |
| Context length | 131,072 | — |
| Licence | apache-2.0 | llama3.2 |
| Commercial use | Allowed | Conditional |
| Languages | — | 8 |
| Downloads 30d | 192,200 | 269,073 |
| Downloads all time | 6.1 M | 15.2 M |
| Quantizations on the Hub | SAFETENSORS | SAFETENSORS |
| Gated | No | Yes — accept licence |
| First seen on the Hub | 2026-04-06 | 2024-09-18 |
Download trend
Daily snapshots, last 90 days (13 Aug – 10 Nov)
4.2 M192 K
granite-4.1-8b
Llama-3.2-3B
VRAM side by side
On RTX 4090 · 24 GB · 8K context unless noted
Llama-3.2-3B · fp16
8.0 GB / 24 GB
granite-4.1-8b · fp16 @ 128K ctx
40.9 GB / 24 GB