QW

Qwen3-Coder-480B-A35B-Instruct

QwenGGUF by unslothUpdated 2025-07-31
Download on Hugging Face
Parameters480B
Context256K
Layers62
Architectureqwen3moe
Smallest quant149.7 GB
Quick answer

Qwen3-Coder-480B-A35B-Instruct needs about 292.6 GB of memory at Q4_K_M (a 290.1 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 23 quants, from 149.7 GB (IQ1_M) to 960.4 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 8.3 GB.

Qwen3-Coder‑480B‑A35B‑Instruct – a large‑scale, agentic code model with 256K native context

Qwen3‑Coder‑480B‑A35B‑Instruct is a 480 B parameter causal language model with 35 B active parameters. It supports 262 144 native context tokens and is optimized for agentic coding and tool‑calling tasks. The model is released under Apache‑2.0 and is available in multiple GGUF quantizations.

Good at

  • Agentic coding and tool calling
  • Large context handling up to 256K tokens
  • High‑quality code generation

Limits

  • Requires high VRAM for full‑precision; no vision support
  • Large file sizes for high‑bit quantizations
  • Context length limited to 262 144 tokens natively

Best for

  • Developing code assistants
  • Repository‑scale code analysis
  • Automated coding workflows

Who it’s for. Engineers and researchers needing a powerful, open‑source code model for large‑scale code generation and analysis. The model is suitable for developers building agentic coding tools and for academic studies of code generation.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_M149.7 GB 152.3 GB
Too big Get file →
Q2_K174.7 GB (4 parts) 177.2 GB
Too big Get file →
Q2_K_L174.9 GB (4 parts) 177.5 GB
Too big Get file →
Q2_K_XL180.3 GB (4 parts) 182.9 GB
Too big Get file →
IQ3_XXS201.5 GB (5 parts) 204.1 GB
Too big Get file →
Q3_K_S206.9 GB (5 parts) 209.5 GB
Too big Get file →
Q3_K_XL212.7 GB (5 parts) 215.2 GB
Too big Get file →
Q3_K_M229.2 GB (5 parts) 231.8 GB
Too big Get file →
IQ4_XS260.5 GB (6 parts) 263.1 GB
Too big Get file →
IQ4_NL270.6 GB (6 parts) 273.2 GB
Too big Get file →
Q4_0271.6 GB (6 parts) 274.2 GB
Too big Get file →
Q4_K_S272.8 GB (6 parts) 275.3 GB
Too big Get file →
Q4_K_XL275.6 GB (6 parts) 278.1 GB
Too big Get file →
Q4_K_M290.1 GB (6 parts) 292.6 GB
Too big Get file →
Q4_1300.5 GB (7 parts) 303.1 GB
Too big Get file →
Q5_K_S330.4 GB (7 parts) 333.0 GB
Too big Get file →
Q5_K_XL340.0 GB (7 parts) 342.6 GB
Too big Get file →
Q5_K_M340.5 GB (7 parts) 343.1 GB
Too big Get file →
Q6_K394.1 GB (9 parts) 396.7 GB
Too big Get file →
Q6_K_XL408.0 GB (9 parts) 410.6 GB
Too big Get file →
Q8_0510.4 GB (11 parts) 512.9 GB
Too big Get file →
Q8_K_XL548.2 GB (12 parts) 550.8 GB
Too big Get file →
BF16960.4 GB (21 parts) 963.0 GB
Too big Get file →

For 8 GB VRAM, use Q4_0 (271.65 GB file, ~274 GB memory at 8k context). For 12 GB VRAM, Q4_K_S (272.75 GB file, ~275 GB memory). For 16 GB VRAM, Q5_K_S (330.44 GB file, ~333 GB memory). File sizes and memory estimates are approximate.

What GPU do you need for Qwen3-Coder-480B-A35B-Instruct?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: no quant fits at an 8K context.
  • 6 GB VRAM: no quant fits at an 8K context.
  • 8 GB VRAM: no quant fits at an 8K context.
  • 12 GB VRAM: no quant fits at an 8K context.
  • 16 GB VRAM: no quant fits at an 8K context.
  • 24 GB VRAM: no quant fits at an 8K context.
  • 32 GB VRAM: no quant fits at an 8K context.
  • 48 GB VRAM: no quant fits at an 8K context.

Which Qwen3-Coder-480B-A35B-Instruct quant should you choose?

Q4_K_M is the usual starting point for Qwen3-Coder-480B-A35B-Instruct: it balances file size and quality well. Here it is a 290.1 GB file.

The Q8_0 file is 510.4 GB, about 1.8 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

If memory is tight, the smallest file is IQ1_M at 149.7 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureqwen3moe
Parameters480B (approximate, mixture of experts)
Active parameters35B
Layers62
Trained context262,144 tokens
KV cache (fp16)248 KB per token

Source

Quantized byunsloth
FamilyQwen
Licenseapache-2.0
Downloads (HF)9,468
Last updated2025-07-31

Frequently asked questions about Qwen3-Coder-480B-A35B-Instruct

How much VRAM does Qwen3-Coder-480B-A35B-Instruct need?

About 292.6 GB at Q4_K_M with an 8K context window. That is the 290.1 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run Qwen3-Coder-480B-A35B-Instruct on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 152.3 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Qwen3-Coder-480B-A35B-Instruct GGUF?

It depends on the quant: IQ1_M is 149.7 GB, Q2_K is 174.7 GB, Q2_K_L is 174.9 GB, Q2_K_XL is 180.3 GB, IQ3_XXS is 201.5 GB, Q3_K_S is 206.9 GB and 17 more.

What context length does Qwen3-Coder-480B-A35B-Instruct support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 2.1 GB at 8K and 8.3 GB at 32K.

Where can I download Qwen3-Coder-480B-A35B-Instruct GGUF?

The files are on Hugging Face in the unsloth/Qwen3-Coder-480B-A35B-Instruct-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does Qwen3-Coder-480B-A35B-Instruct use?

Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run Qwen3-Coder-480B-A35B-Instruct on 8 GB VRAM?

Only with a low‑bit quant like Q4_0; the model will still require ~274 GB of VRAM at 8k context, so it is not feasible on 8 GB VRAM.

Does the model support vision or image inputs?

No, the model has no vision capability and only processes text.

What is the native context length of this model?

The model natively supports 262,144 tokens, extendable up to 1M tokens with Yarn.

Other GGUFs of Qwen3-Coder-480B-A35B-Instruct

Similar models

Browse all →

More from unsloth

All by unsloth →

More Qwen models

All Qwen →

Other large models

All large models →