Qwen3-Coder 30B-A3B
By Alibaba · China
Updated 2026-08-28
Overview
Qwen3-Coder 30B-A3B is Alibaba's MoE coding model with 30B total parameters and 3.3B active, built for agentic coding workflows. Fast on local hardware with native 256K context under Apache 2.0.
When to pick this model
- Local coding copilot for editors like Cline or Aider
- Agentic code workflows needing tool use and fast iteration
- Fast autocomplete or inline suggestions on consumer GPUs
- Commercial products needing a permissive code-model license
VRAM requirements by quantization
| Quantization | VRAM required |
|---|---|
| Q4_K_M (recommended) | 19 GB |
| Q5_K_M | 23 GB |
| Q8_0 | 35 GB |
| FP16 (no quantization) | 61 GB |
VRAM figures include model weights plus a typical 8k KV cache and ~600 MB runtime overhead (Ollama / llama.cpp baseline). Add headroom for higher context lengths.
In practice, Qwen3-Coder 30B-A3B wants a 24 GB card at Q4_K_M (19 GB). Stepping up to Q8_0 nearly doubles the footprint to 35 GB, and unquantized FP16 weights take 61 GB — plan your GPU around the Q4 or Q5 figure unless you specifically need the higher fidelity.
Without a GPU, Qwen3-Coder 30B-A3B needs roughly 32 GB of system RAM to run on CPU via llama.cpp or Ollama — workable for background jobs, but far slower than GPU inference. Throughput estimates from our compatibility engine: around 15 tokens/sec on entry-level GPUs, on the order of 40 tokens/sec on a mid-range card, and up to 100 tokens/sec on high-end hardware — assuming the chosen quantization fully fits in VRAM.
What hardware do you need
The table below matches Qwen3-Coder 30B-A3B to common GPU memory tiers, using the highest-fidelity quantization that fully fits each card class. Spilling layers to system RAM works but costs most of the speed, so size your card to the quantization you actually want to run.
| GPU memory | Example cards | Best fit for Qwen3-Coder 30B-A3B |
|---|---|---|
| 8 GB | RTX 5070 Laptop, RTX 5060, RTX 5060 Ti 8GB | Does not fit — needs 19 GB at Q4_K_M |
| 12 GB | RTX 5070, RTX 5070 Ti Laptop, RTX 4080 Laptop | Does not fit — needs 19 GB at Q4_K_M |
| 16 GB | RTX 5080, RTX 4080 Super, Radeon RX 9070 XT | Does not fit — needs 19 GB at Q4_K_M |
| 24 GB | RTX 4090, Radeon RX 7900 XTX, RTX 5090 Laptop | Q5_K_M (23 GB used) |
| 32 GB | RTX 5090 | Q5_K_M (23 GB used) |
Which GPU should you buy to run Qwen3-Coder 30B-A3B?
To run Qwen3-Coder 30B-A3B locally at Q4, you need ~19 GB of VRAM. The best value for this is a RTX 4090 (24 GB VRAM).
As an Amazon Associate, BestLLMfor earns from qualifying purchases, at no extra cost to you. It does not influence our independent rankings.
Strengths
- 3.3B active parameters — fast inference on local hardware
- Native 262K context
- Apache 2.0, cleared for commercial use
- Official Ollama tag for easy deployment
Limitations
- ~19GB at Q4 — comfortable fit needs 24GB
- Outperformed on benchmarks by GLM-4.7-Flash and Qwen3-Coder-Next
Typical workloads
In our catalog grid, Qwen3-Coder 30B-A3B is filed under Local Copilot, Code Agents, Fast Autocomplete — the use cases where its size/quality trade-off makes the most sense. Its tags translate to concrete workloads: code generation and review (pair it with an editor integration like Continue.dev or Cline).
The 256k-token context window is large enough to hold entire codebases' worth of files or long reports in a single prompt, which is what makes local RAG and document analysis practical. The Apache 2.0 license is permissive, so shipping it inside a commercial product raises no special legal questions.
Architecture & training
Architecture: MoE 30.5B total · 3.3B active · 262K native context
Training: Agentic coding specialist from the Qwen3-Coder family (Alibaba), a smaller sibling to the 480B and Coder-Next models. Trained for tool-augmented use (Cline, Aider).
The easiest-to-deploy agentic coding MoE in the 30B class, though GLM-4.7-Flash now beats it on benchmarks.
Quick start
ollama run qwen3-coder:30bOr use the open-source MCP server to query this model from Claude Desktop, Cursor, or any MCP-compatible client.
Similar models worth comparing
Frequently asked questions
How much VRAM does Qwen3-Coder 30B-A3B need?
At the recommended Q4_K_M quantization, Qwen3-Coder 30B-A3B needs about 19 GB of VRAM. Q8_0 takes 35 GB, and unquantized FP16 weights take 61 GB.
Can Qwen3-Coder 30B-A3B run without a GPU?
Yes — with roughly 32 GB of system RAM it runs CPU-only through llama.cpp or Ollama. Expect a fraction of GPU speed, which is fine for background or batch jobs but slow for interactive chat.
What context window does Qwen3-Coder 30B-A3B support?
Qwen3-Coder 30B-A3B supports a 256k-token context window (262,144 tokens).
Can I use Qwen3-Coder 30B-A3B commercially?
Yes. Qwen3-Coder 30B-A3B is released under Apache 2.0, a permissive open-source license that allows commercial use, modification and redistribution.
How fast is Qwen3-Coder 30B-A3B on consumer hardware?
Our compatibility engine estimates on the order of 40 tokens/sec on a mid-range GPU and up to 100 tokens/sec on high-end cards, assuming the quantization fully fits in VRAM.
Which quantization of Qwen3-Coder 30B-A3B should I download first?
Start with Q4_K_M (19 GB) — the standard size/quality sweet spot. Step up to Q5_K_M or Q8_0 only if you have VRAM headroom. On a 24 GB card you can run up to Q5_K_M.