What are the hardware requirements for Suzume Llama 3 8B Multilingual?

Suzume Llama 3 8B Multilingual requires approximately 4.9 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Suzume Llama 3 8B Multilingual and how large is it?

Suzume Llama 3 8B Multilingual is developed by AlekseyElygin, a model with 8b 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 Suzume Llama 3 8B Multilingual?

Suzume Llama 3 8B Multilingual is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Suzume Llama 3 8B Multilingual by AlekseyElygin — VRAM 4.9GB

Name: Suzume Llama 3 8B Multilingual
Author: AlekseyElygin

Suzume Llama 3 8B Multilingual is an open-source language model by AlekseyElygin. Features: 8b LLM, VRAM: 4.9GB, License: apache-2.0, Quantized, LLM Explorer Score: 0.13.

Base model:lightblue/suzume-ll... Base model:quantized:lightblue... Conversational En Endpoints compatible Gguf Llama Lora Quantized Region:us Trl Unsloth

Model Card on HF 🤗: https://huggingface.co/AlekseyElygin/suzume-llama-3-8B-multilingual

Suzume Llama 3 8B Multilingual Parameters and Internals

Model Type

text-generation-inference, transformers, unsloth, llama, trl

Additional Notes

This llama model was trained 2x faster with Unsloth and Huggingface's TRL library.

LLM Name	Suzume Llama 3 8B Multilingual
Repository 🤗	https://huggingface.co/AlekseyElygin/suzume-llama-3-8B-multilingual
Base Model(s)	Suzume Llama 3 8B Multilingual lightblue/suzume-llama-3-8B-multilingual
Model Size	8b
Required VRAM	4.9 GB
Updated	2026-08-18
Maintainer	AlekseyElygin
Model Files	4.9 GB 5.7 GB 8.5 GB
Supported Languages	en
GGUF Quantization	Yes
Quantization Type	gguf
Model Architecture	AutoModel
License	apache-2.0
Model Max Length	8192
Is Biased	none
Tokenizer Class	PreTrainedTokenizerFast
Padding Token	<\|reserved_special_token_250\|>
PEFT Type	LORA
LoRA Model	Yes
PEFT Target Modules	v_proj\|up_proj\|down_proj\|gate_proj\|o_proj\|k_proj\|q_proj
LoRA Alpha	16
LoRA Dropout	0
R Param	16

Rank the Suzume Llama 3 8B Multilingual 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 Suzume Llama 3 8B Multilingual

Best Alternatives	Context / RAM	Downloads	Likes
...lama 3.1 8B Roleplay BSNL Gguf	0K / 4.9 GB	973	11
...ma 3.1 8B Roleplay Airtel Gguf	0K / 4.9 GB	658	8
Llama 3.1 8B Roleplay Jio Gguf	0K / 4.9 GB	123	5
Llama 3.1 8B Rp Tadashinu Gguf	0K / 4.9 GB	57	7
Llama 3.1 8B OneLastStory Gguf	0K / 4.9 GB	43	5
...B Instruct Bnb 4bit 24 1 100 1	0K / 16.1 GB	59	0
CleverBoi Llama 3.1 8B V2	0K / 16.1 GB	800	0
...leverBoi Llama 3.1 8B Instruct	0K / 0.2 GB	866	1
Meta Llama 3 8B Instruct GGUF	0K / 2 GB	96368	103
...ma 3 8B Instruct 32K V0.1 GGUF	0K / 2 GB	81087	59

Note: green Score (e.g. "73.2") means that the model is better than AlekseyElygin/suzume-llama-3-8B-multilingual.