Qwen3.5-35B-A3B needs about 24.0 GB of memory at its main file (a 22.9 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 3 quants, from 17.0 GB to 25.4 GB. Longer context needs more: at 32K tokens the KV cache alone adds about 2.7 GB.
Qwen3.5-35B-A3B APEX-TQ GGUF model
Qwen3.5-35B-A3B is a 35B parameter causal language model with a mixture of experts. It supports a 262144 token context and is released under the apache-2.0 license. The model is available in experimental APEX-TQ GGUF format.
Good at
- Large context handling
- Mixture-of-Experts efficiency
- Experimental TurboQuant speed
Limits
- Requires TurboQuant Plus fork of llama.cpp
- No standard llama.cpp compatibility
- Large VRAM requirement
Best for
- High‑throughput inference
- Research on MoE models
- Testing TurboQuant optimizations
Who it’s for. Researchers and developers who need a large MoE language model and are willing to use the experimental TurboQuant Plus fork. The model is suitable for those exploring advanced quantization techniques.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 17.0 GB | 18.1 GB | Too big | Get file → | |
| — | 22.9 GB | 24.0 GB | Too big | Get file → | |
| — | 25.4 GB | 26.6 GB | Too big | Get file → |
For 8‑GB VRAM, the 16.98‑GB file is too large; even the 22.88‑GB file exceeds typical VRAM limits. The 25.39‑GB file is the largest and requires more memory. Approximate VRAM usage is 18.1‑26.6 GB at 8 k context, so a GPU with at least 32 GB VRAM is recommended.
What GPU do you need for Qwen3.5-35B-A3B?
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: the best quant that fits is the main file (22.9 GB file, about 24.0 GB in use).
- 32 GB VRAM: the best quant that fits is the main file (25.4 GB file, about 26.6 GB in use).
Which Qwen3.5-35B-A3B quant should you choose?
If memory is tight, the smallest file is the first one in the table at 17.0 GB, with the quality trade-offs that come with fewer bits.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Qwen3.5-35B-A3B
How much VRAM does Qwen3.5-35B-A3B need?
About 24.0 GB at its main file with an 8K context window. That is the 22.9 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 Qwen3.5-35B-A3B on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 18.1 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Qwen3.5-35B-A3B GGUF?
It depends on the quant: file is 17.0 GB, file is 22.9 GB, file is 25.4 GB.
What context length does Qwen3.5-35B-A3B 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 2.7 GB at 32K.
Where can I download Qwen3.5-35B-A3B GGUF?
The files are on Hugging Face in the mudler/Qwen3.5-35B-A3B-APEX-TQ-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 Qwen3.5-35B-A3B 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 8 GB VRAM?
No. The smallest GGUF file requires at least 18.1 GB VRAM at 8 k context, which exceeds 8 GB.
Does this model support vision tasks?
Yes, it is a causal language model with a vision encoder.
What license applies to this model?
The model is released under the apache-2.0 license.