phi 3.1 medium needs about 10.7 GB of memory at Q4_K_M (a 8.6 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 4 quants, from 5.1 GB (Q2_K) to 27.9 GB (F16).
phi 3.1 medium is a 14B-parameter Phi model in GGUF format, quantized by afg1. Its trained context length is 4K tokens. The repository has 4 quant files, from 5.1 GB (Q2_K) to 27.9 GB (F16).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 5.1 GB | 6.5 GB | Fits | Get file → | |
| Q4_K_M | 8.6 GB | 9.9 GB | Too big | Get file → | |
| Q6_K | 11.5 GB | 12.8 GB | Too big | Get file → | |
| F16 | 27.9 GB | 29.3 GB | Too big | Get file → |
What GPU do you need for phi 3.1 medium?
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: the best quant that fits is Q2_K (5.1 GB file, about 7.3 GB in use).
- 12 GB VRAM: the best quant that fits is Q4_K_M (8.6 GB file, about 10.7 GB in use).
- 16 GB VRAM: the best quant that fits is Q6_K (11.5 GB file, about 13.6 GB in use).
- 24 GB VRAM: the best quant that fits is Q6_K (11.5 GB file, about 13.6 GB in use).
- 32 GB VRAM: the best quant that fits is F16 (27.9 GB file, about 30.1 GB in use).
Which phi 3.1 medium quant should you choose?
Q4_K_M is the usual starting point for phi 3.1 medium: it balances file size and quality well. Here it is a 8.6 GB file.
If memory is tight, the smallest file is Q2_K at 5.1 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about phi 3.1 medium
How much VRAM does phi 3.1 medium need?
About 10.7 GB at Q4_K_M with an 8K context window. That is the 8.6 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 3.1 medium on an 8 GB GPU?
Yes, with Q2_K: the 5.1 GB file needs about 7.3 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of phi 3.1 medium GGUF?
It depends on the quant: Q2_K is 5.1 GB, Q4_K_M is 8.6 GB, Q6_K is 11.5 GB, F16 is 27.9 GB.
What context length does phi 3.1 medium support?
The GGUF header lists a trained context length of 4,096 tokens (4K). The KV cache grows with the context you use: about 1.7 GB at 8K and 0.8 GB at 4K.
Where can I download phi 3.1 medium GGUF?
The files are on Hugging Face in the afg1/phi-3.1-medium 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 3.1 medium 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.