PH

Phi 3.5 MoE Instruct

PhiGGUF by phymbertUpdated 2024-12-29
Download on Hugging Face
Parameters41.9B
Context128K
Layers32
Architecturephimoe
Smallest quant18.1 GB
Quick answer

Phi 3.5 MoE Instruct needs about 25.2 GB of memory at Q4_0 (a 23.6 GB file) with an 8K context window, so it fits on a 32 GB GPU. This repository has 4 quants, from 18.1 GB (Q3_K_S) to 83.8 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.

Phi 3.5 MoE Instruct is a 41.9B-parameter Phi model in GGUF format, quantized by phymbert. Its trained context length is 128K tokens. The repository has 4 quant files, from 18.1 GB (Q3_K_S) to 83.8 GB (F16).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q3_K_S18.1 GB 19.6 GB
Too big Get file →
Q4_023.6 GB 25.2 GB
Too big Get file →
Q8_044.5 GB 46.1 GB
Too big Get file →
F1683.8 GB (3 parts) 85.3 GB
Too big Get file →

What GPU do you need for Phi 3.5 MoE 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: the best quant that fits is Q3_K_S (18.1 GB file, about 19.6 GB in use).
  • 32 GB VRAM: the best quant that fits is Q4_0 (23.6 GB file, about 25.2 GB in use).
  • 48 GB VRAM: the best quant that fits is Q8_0 (44.5 GB file, about 46.1 GB in use).

Which Phi 3.5 MoE Instruct quant should you choose?

Q4_0 (23.6 GB) is the middle option in this repository. The usual sweet spot between file size and quality, and the quant most people start with.

If memory is tight, the smallest file is Q3_K_S at 18.1 GB, with the quality trade-offs that come with fewer bits. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturephimoe
Parameters41.9B (approximate, mixture of experts)
Layers32
Trained context131,072 tokens
KV cache (fp16)128 KB per token

Source

Quantized byphymbert
FamilyPhi
Licensemit
Downloads (HF)44
Last updated2024-12-29

Frequently asked questions about Phi 3.5 MoE Instruct

How much VRAM does Phi 3.5 MoE Instruct need?

About 25.2 GB at Q4_0 with an 8K context window. That is the 23.6 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 Phi 3.5 MoE Instruct on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 19.6 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Phi 3.5 MoE Instruct GGUF?

It depends on the quant: Q3_K_S is 18.1 GB, Q4_0 is 23.6 GB, Q8_0 is 44.5 GB, F16 is 83.8 GB.

What context length does Phi 3.5 MoE Instruct support?

The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 1.1 GB at 8K and 4.3 GB at 32K.

Where can I download Phi 3.5 MoE Instruct GGUF?

The files are on Hugging Face in the phymbert/Phi-3.5-MoE-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 Phi 3.5 MoE Instruct use?

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

Similar models

Browse all →

More from phymbert

All by phymbert →

More Phi models

All Phi →

Other large models

All large models →