Qwen3-Coder 30B-A3B (MoE) — VRAM, speed & local setup
Written by Jakub Rusinowski · Last updated
The most-run local coding model. 30.5B total with 3.3B active per token, so it streams like a 3B while reasoning like a 30B, and a 24 GB card holds it comfortably at Q4. 48 layers, 128 experts with 8 active, and a native 262,144-token context.
Qwen3-Coder 30B-A3B (MoE) needs about 19 GB of VRAM at Q4_K_M — quantized weights plus framework overhead, before any KV cache. On Apple Silicon that figure comes out of unified memory.
VRAM and speed by quantization
Quoted against NVIDIA RTX 4090 (24 GB). Includes the KV cache at 8K context, so it reads higher than the headline figure.
| Quant | Memory | VRAM | Speed (est.) | Fit |
|---|---|---|---|---|
| Q2_K 2.63 bpw | 11.6 GB | ~237 tok/s | Fits | |
| Q3_K_M 3.41 bpw | 14.6 GB | ~216 tok/s | Fits | |
| Q4_K_M 4.83 bpw | 20 GB | ~185 tok/s | Fits | |
| Q5_K_M 5.67 bpw | 23.2 GB | ~170 tok/s | Tight | |
| Q6_K 6.56 bpw | 26.6 GB | ~24 tok/s | Offload | |
| Q8_0 8.50 bpw | 34 GB | ~20 tok/s | Offload | |
| F16 16.00 bpw | 62.6 GB | — | Too big |
Black marker = usable memory on the NVIDIA RTX 4090 (24 GB). Estimates from the memory-bandwidth roofline on the methodology page. Qwen3-Coder 30B-A3B (MoE) VRAM calculator →
Get Qwen3-Coder 30B-A3B (MoE) running
The cheapest catalogued GPU that runs Qwen3-Coder 30B-A3B (MoE) is the AMD Radeon RX 7900 XTX (24 GB).
or compare on Vast.ai from $0.35/hr (typical low · varies)
As an Amazon Associate we earn from qualifying purchases. Cloud GPU links are referral links — we may earn a commission at no extra cost to you.
How to run Qwen3-Coder 30B-A3B (MoE)
Install Ollama, then run:
ollama run qwen3-coder:30bSpecifications
Corroborated — Two or more independent sources agree on these figures, but the model card itself was not retrieved. Treat the numbers as good rather than confirmed.
- Parameters
- 30.5 Billion (3.3B active)
- Context window
- 262,144
- Architecture
- MoE
- Provider
- Alibaba Cloud
- Licence
- Apache 2.0
- Practical deployment
- Single consumer GPU
- Specified at
- Q4_K_M
- System RAM
- 32 GB
- Record updated
- 2026-09-19
Commercial use permitted. No usage restrictions beyond attribution.
Quality and use cases
Scores as published by the model’s authors or an independent evaluator — quality, not throughput, and not measured by us.
Can I run Qwen3-Coder 30B-A3B (MoE) on my GPU?
- Qwen3-Coder on AMD Radeon RX 9060 XT 8GB
- Qwen3-Coder on AMD Ryzen AI Max+ 395
- Qwen3-Coder on Apple M2 Max
- Qwen3-Coder on Apple M3 Ultra
- Qwen3-Coder on Intel Arc B570
- Qwen3-Coder on Intel Arc B580
- Qwen3-Coder on NVIDIA A100 80GB (PCIe)
- Qwen3-Coder on NVIDIA GeForce RTX 3060 (12GB)
- Qwen3-Coder on NVIDIA GeForce RTX 3070
- Qwen3-Coder on NVIDIA GeForce RTX 3070 Ti
- Qwen3-Coder on NVIDIA GeForce RTX 3080 (10GB)
- Qwen3-Coder on NVIDIA GeForce RTX 4060
- Qwen3-Coder on NVIDIA GeForce RTX 4070
- Qwen3-Coder on NVIDIA GeForce RTX 4070 Super
- Qwen3-Coder on NVIDIA GeForce RTX 4070 Ti
- Qwen3-Coder on NVIDIA GeForce RTX 5060
- Qwen3-Coder on NVIDIA GeForce RTX 5060 Ti 8GB
- Qwen3-Coder on NVIDIA GeForce RTX 5070
- Qwen3-Coder on NVIDIA H100 80GB (PCIe)
- Qwen3-Coder on NVIDIA RTX PRO 6000 Blackwell
Other Qwen3-Coder sizes
Qwen3-Coder 30B-A3B (MoE) — frequently asked questions
How much VRAM does Qwen3-Coder 30B-A3B (MoE) need?
About 19 GB at Q4_K_M — quantized weights plus framework overhead, before any KV cache. The cache grows with context length and is added on top; the table above folds it in. Apple Silicon counts unified memory toward the same figure.
Does Qwen3-Coder 30B-A3B (MoE) run on an RTX 4090 (24 GB)?
Yes. Qwen3-Coder 30B-A3B (MoE) needs about 19 GB at Q4_K_M, inside a 24 GB card, at an estimated 185 tokens/sec.
How do I run Qwen3-Coder 30B-A3B (MoE) locally?
Install Ollama and run `ollama run qwen3-coder:30b`. That pulls the weights and starts a local OpenAI-compatible endpoint; after the download nothing leaves the machine.
What other sizes does Qwen3-Coder come in?
Qwen3-Coder 8B (6 GB), Qwen3-Coder 30B-A3B (MoE) (19 GB), Qwen3-Coder-Next (80B-A3B MoE) (49 GB), Qwen3-Coder 480B-A35B (MoE) (291 GB). Every size shares the family's training and licence; the larger ones score higher and need proportionally more memory.