NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4 vs Llama-3.1-70B-Instruct

Specs, VRAM requirements and download trends — updated 6 August 2026.

nvidia · A

Params

67.2 B

Context

262,144

30d

2.8 M

meta-llama · B

Params

70.6 B

Context

—

30d

813 K

llama3.1 licence · conditional

Specification comparison

Differences are highlighted; identical values are muted.

Specification	NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4	Llama-3.1-70B-Instruct
Parameters	67.2 B	70.6 B
Architecture	NemotronHForCausalLM	LlamaForCausalLM
Context length	262,144	—
Licence	other	llama3.1
Commercial use	Unknown	Conditional
Languages	7	8
Downloads 30d	2,830,012	813,492
Downloads all time	8.3 M	24.0 M
Quantizations on the Hub	SAFETENSORS	SAFETENSORS
Gated	No	Yes — accept licence
First seen on the Hub	2026-03-10	2024-07-16

Daily snapshots, last 10 days (28 Jul – 6 Aug)

3.1 M807 K

NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4 Llama-3.1-70B-Instruct

On RTX 4090 · 24 GB · 8K context unless noted

Llama-3.1-70B-Instruct · fp16 166.3 GB / 24 GB

❌ Won’t fit

NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4 · fp16 @ 256K ctx 411.6 GB / 24 GB

❌ Won’t fit