Qwen2.5-3B-Instruct vs Qwen3.8-4B-Distill
Specs, VRAM requirements and download trends — updated 25 October 2026.
empero-ai · B · Open weights
Params
4.7 B
Context
—
30d
258 K
apache-2.0 · commercial OK
Specification comparison
Differences are highlighted; identical values are muted.
| Specification | Qwen2.5-3B-Instruct | Qwen3.8-4B-Distill |
|---|---|---|
| Parameters | 3.1 B | 4.7 B |
| Architecture | Qwen2ForCausalLM | Qwen3_5ForConditionalGeneration |
| Context length | 32,768 | — |
| Licence | other | apache-2.0 |
| Commercial use | Unknown | Allowed |
| Languages | 1 | 1 |
| Downloads 30d | 3,718,369 | 258,405 |
| Downloads all time | 124.5 M | 280 K |
| Quantizations on the Hub | SAFETENSORS | SAFETENSORS |
| Gated | No | No |
| First seen on the Hub | 2024-09-17 | 2026-08-15 |
Download trend
Daily snapshots, last 90 days (28 Jul – 25 Oct)
7.6 M165 K
Qwen2.5-3B-Instruct
Qwen3.8-4B-Distill
VRAM side by side
On RTX 4090 · 24 GB · 8K context unless noted
Qwen2.5-3B-Instruct · fp16
7.8 GB / 24 GB
Qwen3.8-4B-Distill · fp16
11.5 GB / 24 GB