What are the hardware requirements for 19B MATH DPO?

19B MATH DPO requires approximately 38.4 GB of VRAM and supports a context window of 4K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed 19B MATH DPO and how large is it?

19B MATH DPO is developed by cloudyu, a model with 19.2b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does 19B MATH DPO perform on standard benchmarks?

19B MATH DPO has the following published scores: MMLU 66.25. Compare against reference models on this page or on the LLM Explorer leaderboards.

19B MATH DPO by cloudyu — VRAM 38.4GB, 4K context

Name: 19B MATH DPO
Author: cloudyu

19B MATH DPO is an open-source language model by cloudyu. Features: 19.2b LLM, VRAM: 38.4GB, Context: 4K, License: other, MoE, LLM Explorer Score: 0.11, Arc: 71.1, HellaSwag: 88.4, MMLU: 66.3, GSM8K: 63.8.

Conversational Dpo Endpoints compatible Mixtral Moe Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/cloudyu/19B_MATH_DPO

19B MATH DPO Parameters and Internals

Model Type

dpo, moe

Additional Notes

This model uses DPO (Direct Preference Optimization) and MoE (Mixture of Experts) with 19 billion parameters.

LLM Name	19B MATH DPO
Repository 🤗	https://huggingface.co/cloudyu/19B_MATH_DPO
Model Size	19.2b
Required VRAM	38.4 GB
Updated	2026-07-29
Maintainer	cloudyu
Model Type	mixtral
Model Files	5.0 GB: 1-of-8 4.9 GB: 2-of-8 5.0 GB: 3-of-8 5.0 GB: 4-of-8 4.9 GB: 5-of-8 5.0 GB: 6-of-8 5.0 GB: 7-of-8 3.6 GB: 8-of-8
Model Architecture	MixtralForCausalLM
License	other
Context Length	4096
Model Max Length	4096
Transformers Version	4.37.2
Tokenizer Class	LlamaTokenizer
Padding Token	<s>
Vocabulary Size	32000
Torch Data Type	bfloat16

Rank the 19B MATH 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

Best Alternatives to 19B MATH DPO

Best Alternatives	Context / RAM	Downloads	Likes
MixTAO 19B Pass	32K / 38.1 GB	3	2
Lorge 2x7B UAMM	32K / 38.2 GB	16	0
Multimerge 19B Pass	32K / 38 GB	10	0
Mistralmath 15B Pass	32K / 38.5 GB	11	0
TaoPassthrough 15B S	32K / 38.4 GB	5	0
Raccoon Small	32K / 38.4 GB	60	1
SPEED 1 BASE	4K / 38.4 GB	172	2
Mixtral 11Bx2 MoE 19B	4K / 38.4 GB	102	38
...oundary Solar Chat 2x10.7B MoE	4K / 38 GB	123	1
SunnyRain 2x10.7B	4K / 38.4 GB	10	0

Note: green Score (e.g. "73.2") means that the model is better than cloudyu/19B_MATH_DPO.