Phi GGUF models

3486 GGUF models in the Phi family, quantized by 1216 uploaders. Sizes run from 1M to 91.7B parameters. The longest trained context is 128K tokens.

TinySmallMediumLarge
Most downloadedNewestLargestLongest contextSmallest
PH
Phi 4 Mini Instruct
Phi · unsloth
Params3.84B
Context128K
Q4_K_M2.5 GB
Coding
PH
Phi 4 Mini Instruct
Phi · bartowski
Params3.84B
Context128K
Q4_K_M2.5 GB
GGUF
PH
Phi 3.5 Mini Instruct
Phi · MaziyarPanahi
Params3.82B
Context128K
Q4_K_M2.4 GB
GGUF
PH
Phi 4 Mini Instruct
Phi · MaziyarPanahi
Params3.84B
Context128K
Q4_K_M2.5 GB
GGUF
PH
phi 4
Phi · MaziyarPanahi
Params14.7B
Context16K
Q4_K_M9.1 GB
GGUF
PH
Phi 3.5 Mini Instruct
Phi · bartowski
Params3.82B
Context128K
Q4_K_M2.4 GB
Coding
PH
Phi 3 mini 4k instruct
Phi · microsoft
Params3.82B
Context4K
Smallest2.4 GB
Coding
PH
phi 2
Phi · TheBloke
Params2.78B
Context2K
Q4_K_M1.8 GB
Coding
PH
phi 4
Phi · microsoft
Params14.7B
Context16K
Smallest3.6 GB
Coding
PH
phi 4
Phi · bartowski
Params14.7B
Context16K
Q4_K_M9.1 GB
Coding
PH
Phi-4-Mini-Reasoning
Phi · unsloth
Params3.84B
Context128K
Q4_K_M2.5 GB
CodingReasoning
PH
Phi 3 mini 4k instruct GGUF Imatrix smashed
Phi · PrunaAI
Params3.82B
Context4K
Q4_K_M2.4 GB
GGUF
PH
Phi 3 mini 4k instruct
Phi · bartowski
Params3.82B
Context4K
Q4_K_M2.4 GB
Coding
PH
Phi-4-Reasoning-Plus
Phi · unsloth
Params14.7B
Context32K
Q4_K_M9.1 GB
CodingReasoning
PH
Phi 3 mini 128k instruct GGUF Imatrix smashed
Phi · PrunaAI
Params3.82B
Context128K
Q4_K_M2.4 GB
GGUF
PH
gretelai Phi 3 mini 128k instruct
Phi · RichardErkhov
Params1.9B
Context—
Q4_K_M0.0 GB
GGUF
PH
phi 4
Phi · lmstudio-community
Params14.7B
Context16K
Q4_K_M9.1 GB
Coding
PH
Phi 3 mini 4k instruct update2024 07 03 IMat
Phi · legraphista
Params3.82B
Context4K
Smallest0.8 GB
Coding
PH
phi3.5 gutenberg 4B
Phi · bartowski
Params3.82B
Context128K
Q4_K_M2.4 GB
GGUF
PH
phi 4
Phi · unsloth
Params14.7B
Context16K
Q4_K_M8.9 GB
Coding
PH
Phi 4 Mini Reasoning
Phi · lmstudio-community
Params3.84B
Context128K
Q4_K_M2.5 GB
CodingReasoning
PH
Phi 3.1 mini 4k instruct
Phi · bartowski
Params3.82B
Context4K
Q4_K_M2.4 GB
Coding
PH
Phi 3.5 Mini instruct_Uncensored
Phi · bartowski
Params3.82B
Context128K
Q4_K_M2.4 GB
GGUF
PH
Phi 3.1 mini 128k instruct
Phi · bartowski
Params3.82B
Context128K
Q4_K_M2.4 GB
Coding
PH
dolphin-2.9.2-Phi-3-Medium
Phi · legraphista
Params14B
Context4K
Smallest3.1 GB
CodingTool use
PH
Phi 4 Mini Instruct
Phi · lmstudio-community
Params3.84B
Context128K
Q4_K_M2.5 GB
GGUF
PH
Phi 3 Medium 128k Instruct
Phi · bartowski
Params14B
Context128K
Q4_K_M8.6 GB
Coding
PH
dolphin-2.9.2-Phi-3-Medium-abliterated
Phi · mradermacher
Params14B
Context4K
Q4_K_M8.4 GB
CodingTool use
PH
Athene Phi 3.5 mini instruct orpo i1
Phi · mradermacher
Params3.82B
Context128K
Q4_K_M2.3 GB
GGUF
PH
Phi 4 Reasoning Plus
Phi · lmstudio-community
Params14.7B
Context32K
Q4_K_M9.1 GB
CodingReasoning

About the Phi GGUF models

Phi is represented on ggufmodels.com by 3,486 GGUF models, quantized by 1,216 uploaders on Hugging Face. The most downloaded are Phi 4 Mini Instruct, Phi 4 Mini Instruct and Phi 3.5 Mini Instruct. Parameter counts run from 1M to 91.7B, and the longest trained context window is 128K tokens.

A Q4_K_M file of a Phi model is between 0.0 to 27.8 GB, which is the quant most people start with. Most files come from mradermacher (660), RichardErkhov (367) and tensorblock (62). The quants you will meet most often are Q4_K_M (2257), Q8_0 (1640), Q5_K_M (1551) and Q6_K (1465), where the number is how many models offer that quant.

Frequently asked questions

Which Phi GGUF should I download?

The most downloaded Phi GGUF here is Phi 4 Mini Instruct. Pick the size that fits your GPU first, then choose a quant: Q4_K_M is the usual starting point, and Q5_K_M or Q6_K give more quality if you have memory to spare. The VRAM pages below list what fits each GPU.

How much VRAM does a Phi model need?

Q4_K_M files of Phi models on this site range from 0.0 to 27.8 GB. Add the KV cache for your context length and about 0.5 GB for the runtime, which usually means 1.5 to 3 GB on top of the file at an 8K context.

Where do Phi GGUF files come from?

The most common quantizers here are mradermacher, RichardErkhov and tensorblock. They convert the original model and publish quantized files on Hugging Face. ggufmodels.com does not host files; every download button opens Hugging Face.

Keep browsing