What are the hardware requirements for DPO Reference Model?

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

Who developed DPO Reference Model and how large is it?

DPO Reference Model is developed by gillesdewaha, a model with 450m 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 DPO Reference Model?

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

DPO Reference Model by gillesdewaha — VRAM 1.8GB

Name: DPO Reference Model
Author: gillesdewaha

DPO Reference Model is an open-source language model by gillesdewaha. Features: 450m LLM, VRAM: 1.8GB, Instruction-Based, LLM Explorer Score: 0.13.

Arxiv:1910.09700 Custom code Instruct Openelm Region:us Safetensors

Model Card on HF 🤗: https://huggingface.co/gillesdewaha/dpo_reference_model

DPO Reference Model Parameters and Internals

LLM Name	DPO Reference Model
Repository 🤗	https://huggingface.co/gillesdewaha/dpo_reference_model
Model Size	450m
Required VRAM	1.8 GB
Updated	2026-07-29
Maintainer	gillesdewaha
Model Type	openelm
Instruction-Based	Yes
Model Files	1.8 GB
Model Architecture	OpenELMForCausalLM
Transformers Version	4.40.2
Tokenizer Class	LlamaTokenizer
Padding Token	</s>
Vocabulary Size	32000
Torch Data Type	float32

Rank the DPO Reference Model 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 DPO Reference Model

Best Alternatives	Context / RAM	Downloads	Likes
OpenELM 450M Instruct	0K / 0.9 GB	798	50
...Databaes SFT DPO 4bitquantized	0K / 0.3 GB	9	0
...450M Instruct Databaes SFT DPO	0K / 0.9 GB	5	0

Note: green Score (e.g. "73.2") means that the model is better than gillesdewaha/dpo_reference_model.