What are the hardware requirements for Ultra Instruct 12B?

Ultra Instruct 12B requires approximately 24.5 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Ultra Instruct 12B and how large is it?

Ultra Instruct 12B is developed by Dampfinchen, a model with 12b 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 Ultra Instruct 12B?

Ultra Instruct 12B is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Ultra Instruct 12B by Dampfinchen — VRAM 24.5GB

Name: Ultra Instruct 12B
Author: Dampfinchen

Ultra Instruct 12B is an open-source language model by Dampfinchen. Features: 12b LLM, VRAM: 24.5GB, License: cc-by-nc-4.0, Instruction-Based, Merged, LLM Explorer Score: 0.14.

Merged Model Arxiv:2306.01708 Arxiv:2311.03099 Base model:intervitensinc/mist... Base model:nbeerbower/mistral-... Base model:pyroserenus/orthrus... Conversational Endpoints compatible Instruct Mistral Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/Dampfinchen/Ultra-Instruct-12B

Ultra Instruct 12B Parameters and Internals

Additional Notes

SMART.

Input Output

Input Format:

Using ChatML

Performance Tips:

Note: It appears it has trouble stopping. If you value extremely long replies, this might be the model for you.

LLM Name	Ultra Instruct 12B
Repository 🤗	https://huggingface.co/Dampfinchen/Ultra-Instruct-12B
Base Model(s)	Pyroserenus/Orthrus-12b-v0.8 Mistral Nemo Gutenberg 12B V2 Deutscher Pantheon 12B Mistral Nemo Base 2407 Chatml Pyroserenus/Orthrus-12b-v0.8 nbeerbower/mistral-nemo-gutenberg-12B-v2 mergekit-community/Deutscher-Pantheon-12B IntervitensInc/Mistral-Nemo-Base-2407-chatml
Merged Model	Yes
Model Size	12b
Required VRAM	24.5 GB
Updated	2026-08-24
Maintainer	Dampfinchen
Instruction-Based	Yes
Model Files	4.9 GB: 1-of-5 4.9 GB: 2-of-5 4.9 GB: 3-of-5 4.9 GB: 4-of-5 4.9 GB: 5-of-5
Model Architecture	AutoModelForCausalLM
License	cc-by-nc-4.0
Tokenizer Class	PreTrainedTokenizerFast
Vocabulary Size	131072

Rank the Ultra Instruct 12B 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 Ultra Instruct 12B

Best Alternatives	Context / RAM	Downloads	Likes
Orthrus 12B V0.8	0K / 24.5 GB	4	4
CleverBoi Nemo 12B V2	0K / 0.2 GB	340	4

Note: green Score (e.g. "73.2") means that the model is better than Dampfinchen/Ultra-Instruct-12B.