by Alibaba (Qwen Team)
Qwen3 4B Base.
text
open weights
edge
4B params
33K ctx
Smallest GPU
1× Nvidia Titan V
Smallest GPU to run it
See all quantisations →
1× Nvidia Titan V.
Most-aggressive quantisation we have a working recommendation for. Lower precision = less VRAM = cheaper hardware, at a small accuracy cost.
Where to use it
Cheapest hosted endpoints.
FAQ
Frequently asked.
How do I run Qwen3 4B Base?
Qwen3 4B Base is open-weight, so you can self-host on rented GPUs. See the Run It Yourself tab for GPU configurations + cost estimates, or use one of the hosted inference providers listed on this page.
Where can I access Qwen3 4B Base?
Qwen3 4B Base is available via Alibaba DashScope, Together AI. Each access option lists its own pricing (per million tokens or hourly hosting).
How much does it cost to run Qwen3 4B Base?
Self-hosting cost depends on the GPU you rent and the throughput you need. See the Run It Yourself tab for GPU configurations and hourly cost estimates.
Is Qwen3 4B Base open-source or proprietary?
Qwen3 4B Base is open-weight under the license. You can download and self-host it.
Run it yourself
Cheapest hardware per quantisation.
Each row is one quantisation tier (the same weights compressed differently). Lower precision → lower VRAM → cheaper hardware, at the cost of small accuracy loss. $/hr refreshed hourly from each provider's API.
| Quantisation | Cheapest GPU config | Total VRAM | Live $/hr | tokens/sec | |
|---|---|---|---|---|---|
|
FP16
FP16 — half precision (default)
|
10 GB | — | — | Compare → | |
|
FP8
FP8 — 8-bit float (Hopper / Blackwell)
|
5 GB | — | — | Compare → | |
|
INT4
INT4 — 4-bit integer (~4× VRAM saving)
|
3 GB | — | — | Compare → |
Just need an API?
Skip the GPU rental and call a hosted endpoint instead.
See access providers on the Overview tab →
Context window
How much it can remember.
33K tokens
≈ 24,576 English words
4K
32K
128K
1M
Capabilities
What it can do.
·
Vision input
·
Audio input
·
Video input
·
Function calling
·
Tool use
·
JSON mode
✓
Streaming
·
Fine-tuning