phi 4 25b i1 needs about 18.7 GB of memory at Q4_K_M (a 15.3 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 23 quants, from 5.4 GB (IQ1_S) to 20.4 GB (Q6_K).
phi 4 25b i1 is a 24.9B-parameter Phi model in GGUF format, quantized by mradermacher. Its trained context length is 16K tokens. The repository has 23 quant files, from 5.4 GB (IQ1_S) to 20.4 GB (Q6_K).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ1_S | 5.4 GB | 8.8 GB | Too big | Get file → | |
| IQ1_M | 5.9 GB | 9.3 GB | Too big | Get file → | |
| IQ2_XXS | 6.7 GB | 10.2 GB | Too big | Get file → | |
| IQ2_XS | 7.5 GB | 10.9 GB | Too big | Get file → | |
| IQ2_S | 7.8 GB | 11.3 GB | Too big | Get file → | |
| IQ2_M | 8.5 GB | 11.9 GB | Too big | Get file → | |
| Q2_K_S | 8.6 GB | 12.0 GB | Too big | Get file → | |
| Q2_K | 9.3 GB | 12.7 GB | Too big | Get file → | |
| IQ3_XXS | 9.8 GB | 13.2 GB | Too big | Get file → | |
| IQ3_XS | 10.4 GB | 13.9 GB | Too big | Get file → | |
| IQ3_S | 10.9 GB | 14.3 GB | Too big | Get file → | |
| Q3_K_S | 10.9 GB | 14.3 GB | Too big | Get file → | |
| IQ3_M | 11.6 GB | 15.0 GB | Too big | Get file → | |
| Q3_K_M | 12.4 GB | 15.8 GB | Too big | Get file → | |
| IQ4_XS | 13.4 GB | 16.8 GB | Too big | Get file → | |
| Q3_K_L | 13.4 GB | 16.8 GB | Too big | Get file → | |
| Q4_0 | 14.2 GB | 17.6 GB | Too big | Get file → | |
| Q4_K_S | 14.2 GB | 17.7 GB | Too big | Get file → | |
| Q4_K_M | 15.3 GB | 18.7 GB | Too big | Get file → | |
| Q4_1 | 15.7 GB | 19.1 GB | Too big | Get file → | |
| Q5_K_S | 17.2 GB | 20.6 GB | Too big | Get file → | |
| Q5_K_M | 18.0 GB | 21.4 GB | Too big | Get file → | |
| Q6_K | 20.4 GB | 23.9 GB | Too big | Get file → |
What GPU do you need for phi 4 25b i1?
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: the best quant that fits is Q2_K_S (8.6 GB file, about 12.0 GB in use).
- 16 GB VRAM: the best quant that fits is Q3_K_M (12.4 GB file, about 15.8 GB in use).
- 24 GB VRAM: the best quant that fits is Q6_K (20.4 GB file, about 23.9 GB in use).
Which phi 4 25b i1 quant should you choose?
Q4_K_M is the usual starting point for phi 4 25b i1: it balances file size and quality well. Here it is a 15.3 GB file.
If memory is tight, the smallest file is IQ1_S at 5.4 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 4 25b i1
How much VRAM does phi 4 25b i1 need?
About 18.7 GB at Q4_K_M with an 8K context window. That is the 15.3 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 4 25b i1 on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 8.8 GB, so you would need a larger GPU or CPU offloading.
What is the file size of phi 4 25b i1 GGUF?
It depends on the quant: IQ1_S is 5.4 GB, IQ1_M is 5.9 GB, IQ2_XXS is 6.7 GB, IQ2_XS is 7.5 GB, IQ2_S is 7.8 GB, IQ2_M is 8.5 GB and 17 more.
What context length does phi 4 25b i1 support?
The GGUF header lists a trained context length of 16,384 tokens (16K). The KV cache grows with the context you use: about 2.9 GB at 8K and 5.9 GB at 16K.
Where can I download phi 4 25b i1 GGUF?
The files are on Hugging Face in the mradermacher/phi-4-25b-i1-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.