Alibaba

Qwen 4B

Alibaba·4B·Qwen·32K ctx·2026-08

Check on my hardwareAdd to compare →Ollama →Hugging Face →

Model size→Hardware requirement→Actual fit

Qwen 4B needs ~3.5–5.6 GB of VRAM depending on quantization. The bar is what it needs; the marker is the card's VRAM. Speed figures below are planning estimates.

8 GB

RTX 4060 · 3060 Ti

5.6
8
Q8_0 · Est. ~14 tok/sComfortable

12 GB

RTX 4070 · 3060 12G

5.6
12
Q8_0 · Est. ~25 tok/sComfortable

16 GB

RTX 4080 · M-series 16G

5.6
16
Q8_0 · Est. ~38 tok/sComfortable

24 GB

RTX 4090 · 3090

5.6
24
Q8_0 · Est. ~59 tok/sComfortable

48 GB

A6000 · 2×24G

5.6
48
Q8_0 · Est. ~59 tok/sComfortable

02  — Quantizations

QuantFile sizeEst. VRAM
Q4_K_M2.3 GB3.5 GB
Q5_K_M2.7 GB3.9 GB
Q8_04.4 GB5.6 GB

03  — Runtimes & use

llama.cpp · Ollama · LM Studio

General · Fast

Memory and speed figures are planning estimates, not measured benchmarks. Memory includes a fixed allowance for KV cache and runtime overhead. Long-context prompts push VRAM use above the numbers shown here.