What are the hardware requirements for Llama 3 Pruned 45B Drobeta Turnu Severin?

Llama 3 Pruned 45B Drobeta Turnu Severin requires approximately 90.2 GB of VRAM and supports a context window of 8K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Llama 3 Pruned 45B Drobeta Turnu Severin and how large is it?

Llama 3 Pruned 45B Drobeta Turnu Severin is developed by Mihaiii, a model with 45b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does Llama 3 Pruned 45B Drobeta Turnu Severin perform on standard benchmarks?

Llama 3 Pruned 45B Drobeta Turnu Severin has the following published scores: MMLU 74.13. Compare against reference models on this page or on the LLM Explorer leaderboards.

Llama 3 Pruned 45B Drobeta Turnu Severin by Mihaiii — VRAM 90.2GB, 8K context

Name: Llama 3 Pruned 45B Drobeta Turnu Severin
Author: Mihaiii

Llama 3 Pruned 45B Drobeta Turnu Severin is an open-source language model by Mihaiii. Features: 45b LLM, VRAM: 90.2GB, Context: 8K, License: other, LLM Explorer Score: 0.14, Arc: 52.9, HellaSwag: 66.7, MMLU: 74.1, GSM8K: 1.1.

Conversational Endpoints compatible Llama Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/Mihaiii/Llama-3-pruned-45B-Drobeta-Turnu-Severin

Llama 3 Pruned 45B Drobeta Turnu Severin Parameters and Internals

Model Type

Transformers, Fine-Tuning

Additional Notes

This model is made with the intention to be used for fine-tuning, not for inference as is. It is a pruned version of the Meta-Llama-3-70B-Instruct model with a reduction in parameter size to approximately 44.9 billion (63% of original size).

LLM Name	Llama 3 Pruned 45B Drobeta Turnu Severin
Repository 🤗	https://huggingface.co/Mihaiii/Llama-3-pruned-45B-Drobeta-Turnu-Severin
Model Size	45b
Required VRAM	90.2 GB
Updated	2026-08-04
Maintainer	Mihaiii
Model Type	llama
Model Files	4.6 GB: 1-of-19 4.7 GB: 2-of-19 5.0 GB: 3-of-19 5.0 GB: 4-of-19 4.7 GB: 5-of-19 4.7 GB: 6-of-19 4.7 GB: 7-of-19 5.0 GB: 8-of-19 5.0 GB: 9-of-19 4.7 GB: 10-of-19 4.7 GB: 11-of-19 4.7 GB: 12-of-19 5.0 GB: 13-of-19 5.0 GB: 14-of-19 4.7 GB: 15-of-19 4.7 GB: 16-of-19 4.7 GB: 17-of-19 5.0 GB: 18-of-19 3.6 GB: 19-of-19
Model Architecture	LlamaForCausalLM
License	other
Context Length	8192
Model Max Length	8192
Transformers Version	4.40.0
Tokenizer Class	PreTrainedTokenizerFast
Vocabulary Size	128256
Torch Data Type	bfloat16

Rank the Llama 3 Pruned 45B Drobeta Turnu Severin 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 Llama 3 Pruned 45B Drobeta Turnu Severin

Best Alternatives	Context / RAM	Downloads	Likes
TinyStories 45M	0.5K / 0.2 GB	58	0

Note: green Score (e.g. "73.2") means that the model is better than Mihaiii/Llama-3-pruned-45B-Drobeta-Turnu-Severin.