AC

ActQuant pi05 3.5bpw

ActquantGGUF by arashakbUpdated 2026-05-13
Download on Hugging Face
Parameters3.39B
Context—
Layers—
Architecturepi05
Smallest quant2.2 GB
Quick answer

ActQuant pi05 3.5bpw needs about 2.9 GB of memory at its main file (a 2.2 GB file) with an 8K context window, so it fits on a 4 GB GPU.

ActQuant pi05 3.5bpw is a 3.39B-parameter Actquant model in GGUF format, quantized by arashakb. The repository has 1 quant file, from 2.2 GB to 2.2 GB.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—2.2 GB 2.9 GB
Fits Get file →

What GPU do you need for ActQuant pi05 3.5bpw?

These figures add the file size, an fp16 KV cache at an 8K context window 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: the best quant that fits is the main file (2.2 GB file, about 2.9 GB in use).

Specifications

Architecture from the GGUF header

Architecturepi05
Parameters3.39B

Source

Quantized byarashakb
FamilyActquant
Downloads (HF)27
Last updated2026-05-13

Frequently asked questions about ActQuant pi05 3.5bpw

How much VRAM does ActQuant pi05 3.5bpw need?

About 2.9 GB at its main file with an 8K context window. That is the 2.2 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 ActQuant pi05 3.5bpw on an 8 GB GPU?

Yes, with the smaller files: the 2.2 GB file needs about 2.9 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of ActQuant pi05 3.5bpw GGUF?

It depends on the quant: file is 2.2 GB.

Where can I download ActQuant pi05 3.5bpw GGUF?

The files are on Hugging Face in the arashakb/ActQuant_pi05_3.5bpw repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

Similar models

Browse all →

More from arashakb

All by arashakb →

More Actquant models

All Actquant →

Other tiny models

All tiny models →