What are the hardware requirements for Snorkel Mistral PairRM DPO?

Snorkel Mistral PairRM DPO requires approximately 14.4 GB of VRAM and supports a context window of 32K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Snorkel Mistral PairRM DPO and how large is it?

Snorkel Mistral PairRM DPO is developed by snorkelai. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does Snorkel Mistral PairRM DPO perform on standard benchmarks?

Snorkel Mistral PairRM DPO has the following published scores: MMLU 60.85. Compare against reference models on this page or on the LLM Explorer leaderboards.

Snorkel Mistral PairRM DPO by snorkelai — VRAM 14.4GB, 32K context

Name: Snorkel Mistral PairRM DPO
Rating: 3 (2 reviews)
Author: snorkelai

Snorkel Mistral PairRM DPO is an open-source language model by snorkelai. Features: LLM, VRAM: 14.4GB, Context: 32K, License: apache-2.0, LLM Explorer Score: 0.12, Arc: 66, HellaSwag: 85.6, MMLU: 60.9, GSM8K: 36.2.

Arxiv:2305.18290 Arxiv:2306.02561 Arxiv:2312.11456 Arxiv:2401.10020 Conversational Dataset:snorkelai/snorkel-mist... Endpoints compatible Mistral Pytorch Region:us Sharded

Model Card on HF 🤗: https://huggingface.co/snorkelai/Snorkel-Mistral-PairRM-DPO

Snorkel Mistral PairRM DPO Parameters and Internals

Model Type

text-generation

Use Cases

Limitations:

The model is a quick demonstration and does not have any moderation mechanisms.

Additional Notes

For enterprise use cases, additional fine-tuning and alignment are necessary. Interested parties can contact Snorkel AI for specialized reward models.

Training Details

Data Sources:

snorkelai/Snorkel-Mistral-PairRM-DPO-Dataset, UltraFeedback

Methodology:

1. Generate five response variations for each prompt from a subset of 20,000 using the LLM - to start, we used [Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2). 2. Apply [PairRM](https://huggingface.co/llm-blender/PairRM) for response reranking. 3. Update the LLM by applying Direct Preference Optimization (DPO) on the top (chosen) and bottom (rejected) responses. 4. Use this LLM as the base model for the next iteration, repeating three times in total.

Input Output

Input Format:

[INST] {prompt} [/INST]

Accepted Modalities:

text

Performance Tips:

The model is designed for initial trials and may take time initially to activate on Hugging Face endpoint.

Release Notes

Version:

GGUF

Notes:

Model version available from community members.

LLM Name	Snorkel Mistral PairRM DPO
Repository 🤗	https://huggingface.co/snorkelai/Snorkel-Mistral-PairRM-DPO
Required VRAM	14.4 GB
Updated	2026-07-30
Maintainer	snorkelai
Model Type	mistral
Model Files	9.9 GB: 1-of-2 4.5 GB: 2-of-2 0.0 GB
Model Architecture	MistralForCausalLM
License	apache-2.0
Context Length	32768
Model Max Length	32768
Transformers Version	4.34.0
Tokenizer Class	LlamaTokenizer
Padding Token	</s>
Vocabulary Size	32000
Torch Data Type	bfloat16

Rank the Snorkel Mistral PairRM DPO 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

Quantized Models of the Snorkel Mistral PairRM DPO

Model	Likes	Downloads	VRAM
...dle Snorkel Mistral PairRM DPO	0	181	14 GB

Best Alternatives to Snorkel Mistral PairRM DPO

Best Alternatives	Context / RAM	Downloads	Likes
Krutrim 2 Instruct	1000K / 49.3 GB	866	38
Ft V1 Violet	1000K / 24.5 GB	5	0
Tiny Random MistralForCausalLM	128K / 0 GB	154	1
Winterreise M7	32K / 14.4 GB	0	0
Frostwind V2.1 M7	32K / 14.4 GB	0	0
...ydaz Web AI Reasoner BaseModel	32K / 14.4 GB	0	1
MistralLite	32K / 14.4 GB	61777	430
MistralLite	32K / 14.4 GB	2373	437
Mixtral AI Cyber Child	32K / 14.5 GB	14	1
Kheops Textbook Immo2	32K / 14.5 GB	11	0

Note: green Score (e.g. "73.2") means that the model is better than snorkelai/Snorkel-Mistral-PairRM-DPO.