What are the hardware requirements for Phi 2 DPO GPTQ?

Phi 2 DPO GPTQ requires approximately 1.8 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Phi 2 DPO GPTQ and how large is it?

Phi 2 DPO GPTQ is developed by TheBloke, a model with 2.8b parameters. 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 DPO GPTQ?

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

Phi 2 DPO GPTQ by TheBloke — VRAM 1.8GB

Name: Phi 2 DPO GPTQ
Author: TheBloke

Phi 2 DPO GPTQ is an open-source language model by TheBloke. Features: 2.8b LLM, VRAM: 1.8GB, License: other, Quantized, LLM Explorer Score: 0.1.

4-bit Base model:lxuechen/phi-2-dpo Base model:quantized:lxuechen/... Code Custom code En Gptq Model-index Phi-msft Quantized Region:us Safetensors

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

Phi 2 DPO GPTQ Parameters and Internals

Model Type

text-generation, instruction-tuned

Use Cases

Areas:

research

Limitations:

Potentially inaccurate code and facts, Limited scope for code outside common Python packages, Unreliable instruction adherence, Language limitations primarily supporting standard English, Potential societal biases, Toxicity, Verbosity

Additional Notes

Original model for fine-tuning: phi-2-sft

Training Details

Data Sources:

UltraFeedback dataset (10k subset)

Methodology:

Direct preference optimization (DPO)

Context Length:

1024

Responsible Ai Considerations

Fairness:

Results may reflect societal biases

Transparency:

Not specified

Accountability:

Unknown

Mitigation Strategies:

Release for research purposes to improve toxicity reduction post-pretraining

Input Output

Input Format:

### Human: {instruction} ### Assistant:

LLM Name	Phi 2 DPO GPTQ
Repository 🤗	https://huggingface.co/TheBloke/phi-2-dpo-GPTQ
Model Name	Phi 2 DPO
Model Creator	Xuechen Li
Base Model(s)	Phi 2 DPO lxuechen/phi-2-dpo
Model Size	2.8b
Required VRAM	1.8 GB
Updated	2026-06-28
Maintainer	TheBloke
Model Type	phi-msft
Model Files	1.8 GB
Supported Languages	en
GPTQ Quantization	Yes
Quantization Type	gptq
Model Architecture	PhiForCausalLM
License	other
Model Max Length	512
Transformers Version	4.37.0.dev0
Tokenizer Class	CodeGenTokenizer
Padding Token	[PAD]
Vocabulary Size	50296
Torch Data Type	float16
Activation Function	gelu_new
Errors	replace

Rank the Phi 2 DPO GPTQ 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 DPO GPTQ

Best Alternatives	Context / RAM	Downloads	Likes
Phi 2 Quantize Gptq	2K / 1.8 GB	6	2
... 2 Electrical Engineering GPTQ	2K / 1.8 GB	9	7
Phi2 OSST GPTQ	2K / 1.8 GB	9	1
Phi 2 Orange GPTQ	0K / 1.8 GB	17	4
Phi 2 GPTQ	0K / 1.8 GB	65	30
Dolphin 2 6 Phi 2 GPTQ	0K / 1.8 GB	5	2
LMCocktail Phi 2 V1 GPTQ	0K / 1.8 GB	6	1
Bnb DPO 8bit	2K / 3 GB	7	0
Bnb DPO 8bit	2K / 3 GB	7	0
Phi 2 4bit 64rank	2K / 5.6 GB	13	0

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