QW

Qwen3.6-35B-A3B-BUDGET-13.5GiB-QAT

QwenGGUF by lmcolemanUpdated 2026-08-20
Download on Hugging Face
Parameters35.5B
Context256K
Layers41
Architectureqwen35moe
Smallest quant14.5 GB
Quick answer

Qwen3.6-35B-A3B-BUDGET-13.5GiB-QAT needs about 15.7 GB of memory at its main file (a 14.5 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 2 quants, from 14.5 GB to 14.5 GB. Longer context needs more: at 32K tokens the KV cache alone adds about 2.8 GB.

Qwen3.6‑35B‑A3B‑BUDGET‑13.5GiB‑QAT – a 35B model compressed to fit 16 GB GPU memory

Qwen3.6‑35B‑A3B‑BUDGET‑13.5GiB‑QAT is a 35.5 B parameter causal language model with 3 B active experts. It supports a native context length of 262,144 tokens and is optimized for 16 GB GPU memory with 64 k context. The model is released under Apache‑2.0.

Good at

  • Reasoning
  • Tools usage
  • Agentic coding

Limits

  • No vision support
  • Requires 16 GB GPU to fit 64 k context
  • Experimental QAT build

Best for

  • Chatbots needing large context
  • Coding assistants
  • Tool‑integrated workflows

Who it’s for. Developers building AI assistants or coding tools that need long‑context reasoning and can allocate 16 GB of GPU memory. The model is suitable for research and production environments that can validate outputs due to its experimental nature.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—14.5 GB 15.7 GB
Too big Get file →
—14.5 GB 15.7 GB
Too big Get file →

The available GGUF file is 14.51 GB and estimates 15.7 GB memory at 8 k context. For a 16 GB GPU, the model fits with 64 k context. If more memory is available, consider the standard MagicQuant hybrids. The file uses an unknown quant; no other quants are listed.

What GPU do you need for Qwen3.6-35B-A3B-BUDGET-13.5GiB-QAT?

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: the best quant that fits is the main file (14.5 GB file, about 15.7 GB in use).

Specifications

Architecture from the GGUF header

Architectureqwen35moe
Parameters35.5B (approximate, mixture of experts)
Active parameters3B
Layers41
Trained context262,144 tokens
KV cache (fp16)82 KB per token

Source

Quantized bylmcoleman
FamilyQwen
Licenseapache-2.0
Downloads (HF)985
Last updated2026-08-20

Frequently asked questions about Qwen3.6-35B-A3B-BUDGET-13.5GiB-QAT

How much VRAM does Qwen3.6-35B-A3B-BUDGET-13.5GiB-QAT need?

About 15.7 GB at its main file with an 8K context window. That is the 14.5 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 Qwen3.6-35B-A3B-BUDGET-13.5GiB-QAT on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 15.7 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Qwen3.6-35B-A3B-BUDGET-13.5GiB-QAT GGUF?

It depends on the quant: file is 14.5 GB, file is 14.5 GB.

What context length does Qwen3.6-35B-A3B-BUDGET-13.5GiB-QAT support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.7 GB at 8K and 2.8 GB at 32K.

Where can I download Qwen3.6-35B-A3B-BUDGET-13.5GiB-QAT GGUF?

The files are on Hugging Face in the lmcoleman/Qwen3.6-35B-A3B-BUDGET-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 Qwen3.6-35B-A3B-BUDGET-13.5GiB-QAT 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 this model on 16 GB VRAM?

Yes, the model plus 64 k context fits in 16 GB of GPU memory as measured on unified memory.

What does QAT mean for this model?

QAT indicates the model was trained with quantization‑aware training to preserve perplexity after aggressive low‑bit quantization.

Under what license is the model released?

The model is released under the Apache‑2.0 license.

Other GGUFs of Qwen3.6-35B-A3B

Similar models

Browse all →

More from lmcoleman

All by lmcoleman →

More Qwen models

All Qwen →

Other large models

All large models →