llama-3.3-70b-instruct-awq vs MiniMax-M2.5-NVFP4

Specs, VRAM requirements and download trends — updated 31 August 2026.

casperhansen · A
Params
70.6 B
Context
131,072
30d
507 K
llama3.3 licence · conditional
nvidia · B
Params
116.4 B
Context
196,608
30d
222 K

Specification comparison

Differences are highlighted; identical values are muted.

Specification llama-3.3-70b-instruct-awq MiniMax-M2.5-NVFP4
Parameters 70.6 B 116.4 B
Architecture LlamaForCausalLM MiniMaxM2ForCausalLM
Context length 131,072 196,608
Licence llama3.3 other
Commercial use Conditional Unknown
Languages 8
Downloads 30d 507,212 221,989
Downloads all time 4.4 M 1.1 M
Quantizations on the Hub SAFETENSORS SAFETENSORS
Gated No No
First seen on the Hub 2024-12-06 2026-03-11

Download trend

Daily snapshots, last 56 days (28 Jul – 21 Sep)

822 K222 K
llama-3.3-70b-instruct-awq MiniMax-M2.5-NVFP4 A overtook B on 13 Sep

VRAM side by side

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

llama-3.3-70b-instruct-awq · fp16 @ 128K ctx 213.6 GB / 24 GB
❌ Won’t fit
MiniMax-M2.5-NVFP4 · fp16 @ 192K ctx 573.2 GB / 24 GB
❌ Won’t fit

Adjacent comparisons