Qwen2.5-7B-Instruct vs Qwen3.6-35B-A3B-NVFP4

Specs, VRAM requirements and download trends — updated 20 September 2026.

Qwen · A
Params
7.6 B
Context
32,768
30d
9.7 M
apache-2.0 · commercial OK
nvidia · B
Params
18.7 B
Context
30d
8.3 M
apache-2.0 · commercial OK

Specification comparison

Differences are highlighted; identical values are muted.

Specification Qwen2.5-7B-Instruct Qwen3.6-35B-A3B-NVFP4
Parameters 7.6 B 18.7 B
Architecture Qwen2ForCausalLM Qwen3_5MoeForConditionalGeneration
Context length 32,768
Licence apache-2.0 apache-2.0
Commercial use Allowed Allowed
Languages 1
Downloads 30d 9,726,568 8,277,959
Downloads all time 182.2 M 34.5 M
Quantizations on the Hub SAFETENSORS SAFETENSORS
Gated No No
First seen on the Hub 2024-09-16 2026-05-27

Download trend

Daily snapshots, last 55 days (28 Jul – 20 Sep)

12.6 M8.3 M
Qwen2.5-7B-Instruct Qwen3.6-35B-A3B-NVFP4 A overtook B on 4 Sep

VRAM side by side

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

Qwen2.5-7B-Instruct · fp16 18.4 GB / 24 GB
✅ Runs comfortably
Qwen3.6-35B-A3B-NVFP4 · fp16 29.1 GB / 24 GB
❌ Won’t fit

Adjacent comparisons