PX

PXA Fusion4 35B

PxaGGUF by poisonxaUpdated 2026-10-12
Download on Hugging Face
Parameters35.5B
Context256K
Layers41
Architectureqwen35moe
Smallest quant11.2 GB
Quick answer

PXA Fusion4 35B needs about 12.4 GB of memory at its main file (a 11.2 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 2 quants, from 11.2 GB to 19.3 GB. Longer context needs more: at 32K tokens the KV cache alone adds about 2.8 GB.

PXA Fusion4 35B – 35B Mixture-of-Experts model tuned for agentic tool-use and reasoning

PXA Fusion4 35B is a 35.5B parameter Mixture-of-Experts model based on Qwen3.5-35B-A3B. It is optimized for agentic tool-use, coding, and uncensored reasoning. The model runs fully GPU‑resident on modest hardware via PXQ quantization.

Good at

  • Agentic tool‑use
  • Coding tasks
  • Uncensored reasoning

Limits

  • Requires 8k context for full memory estimate
  • No vision support
  • License: apache-2.0

Best for

  • Building autonomous agents
  • Writing code snippets
  • Long‑form reasoning

Who it’s for. Developers and researchers who need a large‑scale, agentic LLM that can run on single or dual 16‑GB GPUs. The model is suitable for projects requiring reasoning and tool‑use without fabrication.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—11.2 GB 12.4 GB
Too big Get file →
—19.3 GB 20.5 GB
Too big Get file →

For a single 16‑GB GPU, use the 11.24‑GB file (estimated 12.4‑GB memory at 8k context). For a single 24‑GB GPU or dual 16‑GB GPUs, use the 19.3‑GB file (estimated 20.5‑GB memory at 8k context). Memory estimates are approximate.

What GPU do you need for PXA Fusion4 35B?

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 (11.2 GB file, about 12.4 GB in use).
  • 24 GB VRAM: the best quant that fits is the main file (19.3 GB file, about 20.5 GB in use).

Specifications

Architecture from the GGUF header

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

Source

Quantized bypoisonxa
FamilyPxa
Licenseapache-2.0
Downloads (HF)1,131
Last updated2026-10-12

Frequently asked questions about PXA Fusion4 35B

How much VRAM does PXA Fusion4 35B need?

About 12.4 GB at its main file with an 8K context window. That is the 11.2 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 PXA Fusion4 35B on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 12.4 GB, so you would need a larger GPU or CPU offloading.

What is the file size of PXA Fusion4 35B GGUF?

It depends on the quant: file is 11.2 GB, file is 19.3 GB.

What context length does PXA Fusion4 35B 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 PXA Fusion4 35B GGUF?

The files are on Hugging Face in the poisonxa/PXA-Fusion4-35B-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 PXA Fusion4 35B 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 PXA Fusion4 35B on 8 GB VRAM?

No, the smallest quant requires at least 12.4 GB of VRAM at 8k context. 8 GB is insufficient.

What quantization is used for this model?

The model uses PXQ quantization, available in two tiers: a 11.24‑GB tier for 16‑GB GPUs and a 19.3‑GB tier for 24‑GB or dual 16‑GB GPUs.

Does the model support vision?

No, the model does not support vision capabilities.

Similar models

Browse all →

More from poisonxa

All by poisonxa →

More Pxa models

All Pxa →

Other large models

All large models →