What are the hardware requirements for Ultiima 78B?

Ultiima 78B requires approximately 156.7 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 Ultiima 78B and how large is it?

Ultiima 78B is developed by Sakalti, a model with 78b 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 Ultiima 78B?

Ultiima 78B is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Ultiima 78B by Sakalti — VRAM 156.7GB, 32K context

Name: Ultiima 78B
Author: Sakalti

Ultiima 78B is an open-source language model by Sakalti. Features: 78b LLM, VRAM: 156.7GB, Context: 32K, License: other, Instruction-Based, Merged, LLM Explorer Score: 0.16.

Merged Model Arxiv:2306.01708 Base model:maziyarpanahi/calme... Base model:saxo/linkbricks-hor... Conversational Endpoints compatible Instruct Qwen2 Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/Sakalti/ultiima-78B

Ultiima 78B Parameters and Internals

LLM Name	Ultiima 78B
Repository 🤗	https://huggingface.co/Sakalti/ultiima-78B
Base Model(s)	...cks Horizon AI Avengers V2 78B MaziyarPanahi/calme-3.2-instruct-78b Saxo/Linkbricks-Horizon-AI-Avengers-V2-78B MaziyarPanahi/calme-3.2-instruct-78b
Merged Model	Yes
Model Size	78b
Required VRAM	156.7 GB
Updated	2026-07-26
Maintainer	Sakalti
Model Type	qwen2
Instruction-Based	Yes
Model Files	5.0 GB: 1-of-33 5.0 GB: 2-of-33 4.8 GB: 3-of-33 4.8 GB: 4-of-33 4.8 GB: 5-of-33 5.0 GB: 6-of-33 4.8 GB: 7-of-33 4.8 GB: 8-of-33 4.8 GB: 9-of-33 5.0 GB: 10-of-33 4.8 GB: 11-of-33 4.8 GB: 12-of-33 4.8 GB: 13-of-33 5.0 GB: 14-of-33 4.8 GB: 15-of-33 4.8 GB: 16-of-33 4.8 GB: 17-of-33 5.0 GB: 18-of-33 4.8 GB: 19-of-33 4.8 GB: 20-of-33 4.8 GB: 21-of-33 5.0 GB: 22-of-33 4.8 GB: 23-of-33 4.8 GB: 24-of-33 4.8 GB: 25-of-33 5.0 GB: 26-of-33 4.8 GB: 27-of-33 4.8 GB: 28-of-33 4.8 GB: 29-of-33 5.0 GB: 30-of-33 4.8 GB: 31-of-33 4.8 GB: 32-of-33 1.3 GB: 33-of-33
Model Architecture	Qwen2ForCausalLM
License	other
Context Length	32768
Model Max Length	32768
Transformers Version	4.46.2
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Vocabulary Size	151646
Torch Data Type	float16
Errors	replace

Rank the Ultiima 78B 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 Ultiima 78B

Best Alternatives	Context / RAM	Downloads	Likes
Calme 3.2 Instruct 78B	32K / 207.1 GB	755	223
Calme 3.1 Instruct 78B	32K / 207.1 GB	67	8
BBAI 78B Calme 3 1 Ties	32K / 54.4 GB	7	0
Calme Ties2 78B	32K / 156.7 GB	6	1
Calme Ties 78B	32K / 156.7 GB	4	1
...cks Horizon AI Avengers V2 78B	32K / 156.7 GB	5	0
...cks Horizon AI Avengers V1 78B	32K / 156.7 GB	6	0

Note: green Score (e.g. "73.2") means that the model is better than Sakalti/ultiima-78B.