v2.2 · 2026-06-12

Qwen3-Coder 30B A3B Instruct

qwen3-coder-30b-a3b-instruct

CODE

Qwen3-Coder 30B A3B Instruct is a 30.5B code model profile for local AI planning. This page records provider, license, context, quantization, hardware-fit estimates, setup hints, source links, and caveats so readers do not choose a model by name alone.

PROVIDER

Alibaba Qwen

FAMILY

Qwen3-Coder

MODEL TYPE

code

PARAMETERS

30.5B (3.3B active)

MODALITIES

text, code

ARCHITECTURE

Mixture-of-Experts causal language model; 128 experts, 8 activated; grouped-query attention; non-thinking instruct model

CONTEXT WINDOW

262,144 tokens

TRAINING TOKENS

—

RELEASE DATE

2025-07

LICENSE & LINKS

Apache-2.0license text

MODEL CARD

huggingface.co/Qwen/Qwen3-Coder-30B-A3B-Instruct

HUGGING FACE

Qwen/Qwen3-Coder-30B-A3B-Instruct

GGUF REPOSITORIES

unsloth/Qwen3-Coder-30B-A3B-Instruct-GGUF community conversion

Community GGUF repository referenced by the June 2026 model research.

SETUP HINTS

OLLAMA

ollama run qwen3-coder:30b

LM STUDIO

Available according to the supplied June 2026 research; select a cited GGUF/MLX build and verify current app metadata before relying on hands-on setup guidance.

LLAMA.CPP

Use the cited GGUF repository (unsloth/Qwen3-Coder-30B-A3B-Instruct-GGUF) with current llama.cpp-compatible tooling.

QUANTIZATION FILE SIZES (GGUF)

Q4_K_M

18.6 GB

Q5_K_M

21.7 GB

Q8_0

32.5 GB

Quantization note: Q4_K_M for first local attempts unless a cited runtime recommends a different default.

RAM / VRAM ESTIMATES

MIN RAM

15 GB

COMFORTABLE RAM

32 GB

MIN VRAM

15 GB

COMFORTABLE VRAM

32 GB

These are conservative local-inference estimates, not official hardware requirements.

HARDWARE FIT

CPU only (no GPU)

limited

8 GB RAM

16 GB RAM

limited

32 GB RAM

comfortable

8 GB VRAM

12 GB VRAM

24 GB VRAM

limited

Apple Silicon (unified memory)

limited

Hardware fit values are conservative local-inference estimates based on GGUF size plus runtime and KV-cache overhead. Actual requirements depend on context length, quantization, runtime, GPU offload, and other running apps.

BEST FOR

·Developers wanting local repo help, code completion, debugging, and lightweight agent workflows on serious laptops.

AVOID IF

·You need tiny RAM usage, fast CPU only responses, or reliable full 256K contexts locally.

CAVEATS

·Exact standalone release date for the 30B checkpoint was not confirmed from a primary Qwen announcement in this pass.

·Full 256K context can require far more memory than the model file alone.

·Function calling depends on current runtime support for the qwen3_coder tool parser or compatible chat templates.

·Use updated llama.cpp, Ollama, LM Studio, vLLM, or SGLang builds if tool calling behaves incorrectly.

·Non-thinking model; users should not expect visible chain-of-thought style output.

·Commercial-use status from the supplied research: Permitted under Apache-2.0 terms

·Beginner summary from supplied research: Qwen3 Coder 30B is the practical local coding choice: it fits high memory laptops, understands large repositories, supports completion and tool workflows, and is much easier to run than 80B or 480B alternatives locally today.

SOURCE URLS