v2.2 · 2026-06-12

Qwen3-Coder 480B A35B Instruct

qwen3-coder-480b-a35b-instruct

CODE

Qwen3-Coder 480B A35B Instruct is a 480B 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

480B (35B active)

MODALITIES

text, code

ARCHITECTURE

Mixture-of-Experts causal language model; 160 experts, 8 activated; grouped-query attention; non-thinking instruct model

CONTEXT WINDOW

262,144 tokens

TRAINING TOKENS

—

RELEASE DATE

2025-07-22

LICENSE & LINKS

Apache-2.0license text

MODEL CARD

huggingface.co/Qwen/Qwen3-Coder-480B-A35B-Instruct

HUGGING FACE

Qwen/Qwen3-Coder-480B-A35B-Instruct

GGUF REPOSITORIES

unsloth/Qwen3-Coder-480B-A35B-Instruct-GGUF community conversion

Community GGUF repository referenced by the June 2026 model research.

SETUP HINTS

OLLAMA

ollama run qwen3-coder:480b

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-480B-A35B-Instruct-GGUF) with current llama.cpp-compatible tooling.

QUANTIZATION FILE SIZES (GGUF)

Q4_K_M

276 GB

Q5_K_M

339 GB

Q8_0

510 GB

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

RAM / VRAM ESTIMATES

MIN RAM

250 GB

COMFORTABLE RAM

384 GB

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

HARDWARE FIT

CPU only (no GPU)

8 GB RAM

16 GB RAM

32 GB RAM

8 GB VRAM

12 GB VRAM

24 GB VRAM

Apple Silicon (unified memory)

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

·Benchmark chasing, server side code agents, and maximum open model coding quality experiments at scale.

AVOID IF

·You want everyday local coding, affordable hardware, quick downloads, or simple laptop deployment at home.

CAVEATS

·This should not be presented as a normal beginner local model.

·Even quantized builds are hundreds of gigabytes before runtime overhead and context memory.

·Ollama and LM Studio availability does not mean typical consumer hardware can run it well.

·Full 256K context and agentic workflows require substantially more memory than the model file size.

·Non-thinking model; it is optimized for coding and agentic tasks, not visible reasoning traces.

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

·Beginner summary from supplied research: Qwen3 Coder 480B is Qwen’s flagship open coding model for agentic software engineering, but local use belongs to servers, multi GPU rigs, or huge unified memory machines rather than normal laptops for daily development work.

SOURCE URLS