What are the hardware requirements for Phi 1 5 GGUF?

Phi 1 5 GGUF requires approximately 0.6 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Phi 1 5 GGUF and how large is it?

Phi 1 5 GGUF is developed by TKDKid1000. 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 1 5 GGUF?

Phi 1 5 GGUF is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Phi 1 5 GGUF by TKDKid1000 — VRAM 0.6GB

Name: Phi 1 5 GGUF
Author: TKDKid1000

Phi 1 5 GGUF is an open-source language model by TKDKid1000. Features: LLM, VRAM: 0.6GB, License: other, Quantized, LLM Explorer Score: 0.11.

Arxiv:2309.05463 Code Custom code En Gguf Phi-msft Pytorch Q2 Quantized Region:us

Model Card on HF 🤗: https://huggingface.co/TKDKid1000/phi-1_5-GGUF

Phi 1 5 GGUF Parameters and Internals

Model Type

text generation, code generation

Use Cases

Areas:

research

Applications:

poem writing, email drafting, story creation, text summarization, Python code generation

Limitations:

produces incorrect code snippets and statements, to be treated as suggestions, not fine-tuned for instructions, language limitations to standard English only

Considerations:

For research purposes; not suggested for direct adoption in production

Additional Notes

Trained with selected safe data, but still vulnerable to producing harmful content.

Supported Languages

en (standard English)

Training Details

Data Sources:

same sources used for phi-1, with additional NLP synthetic texts

Data Volume:

150 billion tokens

Training Time:

8 days

Hardware Used:

32xA100-40G GPUs

Model Architecture:

Transformer with next-word prediction objective

Input Output

Input Format:

QA, chat, and code formats

Accepted Modalities:

text

Output Format:

QA, chat, and code formats

Performance Tips:

Use FP16 / CUDA for optimal performance

Release Notes

Version:

initial

Notes:

Model provides near state-of-the-art performance in language tasks for its parameter size, not fine-tuned for instructions.

LLM Name	Phi 1 5 GGUF
Repository 🤗	https://huggingface.co/TKDKid1000/phi-1_5-GGUF
Required VRAM	0.6 GB
Updated	2026-05-13
Maintainer	TKDKid1000
Model Type	phi-msft
Model Files	0.6 GB 0.8 GB 0.9 GB 1.1 GB 1.2 GB 1.5 GB 2.8 GB 2.8 GB
Supported Languages	en
GGUF Quantization	Yes
Quantization Type	gguf\|q2\|q4_k\|q5_k
Model Architecture	PhiForCausalLM
License	other
Model Max Length	2048
Transformers Version	4.34.1
Tokenizer Class	CodeGenTokenizer
Vocabulary Size	51200
Torch Data Type	float16
Activation Function	gelu_new

Rank the Phi 1 5 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 1 5 GGUF

Best Alternatives	Context / RAM	Downloads	Likes
MFANN3bv0.7.10 GGUF	2K / 1.6 GB	8	0
Candle Phi 1	2K / 2.8 GB	84	0
Candle Phi 1 5	2K / 2.8 GB	68	0
Phi 2 GGUF	2K / 1.1 GB	383	3
Dolphin 2 6 Phi 2 Candle	0K / 0 GB	250	5
Phi 2 Quantized	0K / 1.6 GB	114	11
Phi 1 5 Q4	2K / GB	26	1
Phi 2 Hf 4bit Mlx	2K / 1.8 GB	65	2
Merlin1.4	2K / 5.6 GB	74	0
Merlin1.5	2K / 5.6 GB	99	0

Note: green Score (e.g. "73.2") means that the model is better than TKDKid1000/phi-1_5-GGUF.