Configure a rig →

Qwen4B paramspopular

Qwen3 4B Base locally

Qwen3 4B Base is an open-weight LLM from the Qwen family with 4B parameters. Main use: chat, RAG and general assistance. Detected minimum hardware: GTX 1650 (4 GB).

Check on my GPU →Hugging Face ↗

Technical facts

Parameters4B

Q4_K_M2.5 GB

Q5_K_M3.1 GB

Q84.5 GB

FP168.9 GB

FamilyQwen

Last sync2026-05-12

Available quantizations

GGUF weights

Q4_K_M

2.5GB

Acceptable. Good compromise when VRAM is limited.

Q5_K_M

3.1GB

Good quality. Sweet spot for size and precision.

Q8

4.5GB

Near-FP16 quality. Comfortable for production.

FP16

8.9GB

Reference precision. Maximum quality, doubled VRAM.

Compatible GPUs

12 single-GPU

GPUs that can run Qwen3 4B Base on a single card, ranked by VRAM headroom.

3.1 / 4 GBcomfortable · Q5

4.5 / 6 GBcomfortable · Q8

4.5 / 6 GBcomfortable · Q8

4.5 / 6 GBcomfortable · Q8

4.5 / 6 GBcomfortable · Q8

4.5 / 6 GBcomfortable · Q8

4.5 / 6 GBcomfortable · Q8

6 GB · Arc Alchemist

4.5 / 6 GBcomfortable · Q8

4.5 / 8 GBcomfortable · Q8

4.5 / 8 GBcomfortable · Q8

4.5 / 8 GBcomfortable · Q8

4.5 / 8 GBcomfortable · Q8

Recommended multi-GPU rigs

2x / 4x consumer GPUs

For Qwen3 4B Base at higher quantization or with more context, a multi-GPU rig gives more headroom.

4.5 / 8 GBcomfortable · Q8

2× GTX 1060 6GB

12 GB · GTX 10

8.9 / 12 GBcomfortable · FP16

12 GB · GTX 16

8.9 / 12 GBcomfortable · FP16

2× GTX 1660 Super

12 GB · GTX 16

8.9 / 12 GBcomfortable · FP16

2× GTX 1660 Ti

12 GB · GTX 16

8.9 / 12 GBcomfortable · FP16

2× RTX 2060 6GB

12 GB · RTX 20

8.9 / 12 GBcomfortable · FP16

2× RTX 3050 6GB

12 GB · RTX 30

8.9 / 12 GBcomfortable · FP16

12 GB · Arc Alchemist

8.9 / 12 GBcomfortable · FP16

Recommended rig

2× GTX 1650

Qwen3 4B Base with Ubuntu, vLLM, Open WebUI and the model already downloaded.

Similar models

VRAM estimates: parameters x bits/8 plus margin. Real performance varies by engine, context length and batch size.
sync: 2026-05-12