What are the hardware requirements for Phi 2 GGUF?

Phi 2 GGUF requires approximately 1.2 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Phi 2 GGUF and how large is it?

Phi 2 GGUF is developed by TheBloke. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Phi 2 GGUF?

Phi 2 GGUF is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Phi 2 GGUF by TheBloke — VRAM 1.2GB

Name: Phi 2 GGUF
Author: TheBloke

Phi 2 GGUF is an open-source language model by TheBloke. Features: LLM, VRAM: 1.2GB, License: other, Quantized, LLM Explorer Score: 0.15.

Base model:microsoft/phi-2 Base model:quantized:microsoft... Code En Gguf Phi-msft Quantized Region:us

Model Card on HF 🤗: https://huggingface.co/TheBloke/phi-2-GGUF

Phi 2 GGUF Parameters and Internals

Model Type

phi-msft, text-generation

Use Cases

Areas:

research

Limitations:

Generate Inaccurate Code and Facts, Unreliable Responses to Instruction, Language Limitations, Potential Societal Biases, Toxicity, Verbosity

Additional Notes

Phi-2 is intended for research purposes only and has not been fine-tuned through reinforcement learning from human feedback.

Supported Languages

en (standard)

Training Details

Data Sources:

NLP synthetic data created by AOAI GPT-3.5, filtered web data from Falcon RefinedWeb and SlimPajama

Data Volume:

250B tokens

Context Length:

2048

Training Time:

14 days

Hardware Used:

96xA100-80G

Model Architecture:

Transformer-based model with next-word prediction objective

Input Output

Input Format:

Instruct: {prompt}\nOutput:

Accepted Modalities:

text

Output Format:

text

LLM Name	Phi 2 GGUF
Repository 🤗	https://huggingface.co/TheBloke/phi-2-GGUF
Model Name	Phi 2
Model Creator	Microsoft
Base Model(s)	Phi 2 microsoft/phi-2
Required VRAM	1.2 GB
Updated	2026-07-26
Maintainer	TheBloke
Model Type	phi-msft
Model Files	1.2 GB 1.6 GB 1.5 GB 1.2 GB 1.6 GB 1.8 GB 1.6 GB 1.9 GB 2.1 GB 1.9 GB 2.3 GB 3.0 GB
Supported Languages	en
GGUF Quantization	Yes
Quantization Type	gguf
Model Architecture	AutoModel
License	other

Rank the Phi 2 GGUF capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Best Alternatives to Phi 2 GGUF

Best Alternatives	Context / RAM	Downloads	Likes
ComicBot V.2 Gguf	32K / 5 GB	54	0
Gemma4 E2 Code Lora Gguf	0K / 0.1 GB	110	1
Gemma4 E2B Coding Distill	0K / 0.2 GB	44	1
Qwen3 Medical GRPO GGUF	0K / 1.7 GB	440	6
Gemma2 WizardLM	0K / 5.2 GB	22	0
...ixtral 8x7B Instruct V0.1 GGUF	0K / 15.6 GB	18491	655
Marco O1 GGUF	0K / 3 GB	262	6
Dolphin 2.5 Mixtral 8x7b GGUF	0K / 15.6 GB	18101	314
Dolphin 2.7 Mixtral 8x7b GGUF	0K / 15.6 GB	12812	176
Mixtral 8x7B V0.1 GGUF	0K / 15.6 GB	4192	438

Note: green Score (e.g. "73.2") means that the model is better than TheBloke/phi-2-GGUF.