What are the hardware requirements for EEVE Korean Instruct 10.8B Var?

EEVE Korean Instruct 10.8B Var requires approximately 21.6 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 EEVE Korean Instruct 10.8B Var and how large is it?

EEVE Korean Instruct 10.8B Var is developed by maywell, a model with 10.8b 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 EEVE Korean Instruct 10.8B Var?

EEVE Korean Instruct 10.8B Var is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

EEVE Korean Instruct 10.8B Var by maywell — VRAM 21.6GB, 4K context

Name: EEVE Korean Instruct 10.8B Var
Author: maywell

EEVE Korean Instruct 10.8B Var is an open-source language model by maywell. Features: 10.8b LLM, VRAM: 21.6GB, Context: 4K, License: apache-2.0, Instruction-Based, LLM Explorer Score: 0.12.

Conversational Endpoints compatible Instruct Llama Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/maywell/EEVE-Korean-Instruct-10.8B-Var

EEVE Korean Instruct 10.8B Var Parameters and Internals

Additional Notes

Created by tuning only layers 46 and 47.

Training Details

Methodology:

The model is created by fine-tuning layers 46 and 47.

LLM Name	EEVE Korean Instruct 10.8B Var
Repository 🤗	https://huggingface.co/maywell/EEVE-Korean-Instruct-10.8B-Var
Model Size	10.8b
Required VRAM	21.6 GB
Updated	2026-08-07
Maintainer	maywell
Model Type	llama
Instruction-Based	Yes
Model Files	4.9 GB: 1-of-5 4.9 GB: 2-of-5 5.0 GB: 3-of-5 4.9 GB: 4-of-5 1.9 GB: 5-of-5
Model Architecture	LlamaForCausalLM
License	apache-2.0
Context Length	4096
Model Max Length	4096
Transformers Version	4.40.0.dev0
Tokenizer Class	LlamaTokenizer
Padding Token	</s>
Vocabulary Size	40960
Torch Data Type	bfloat16

Rank the EEVE Korean Instruct 10.8B Var 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 EEVE Korean Instruct 10.8B Var

Best Alternatives	Context / RAM	Downloads	Likes
...Korean Instruct 10.8B V1.0 32K	32K / 21.6 GB	40	10
...EVE Korean Instruct 10.8B V1.0	4K / 21.6 GB	9491	160
...anoljaNEXT EEVE Instruct 10.8B	4K / 21.6 GB	14441	166
...Judgment Transducer 10.8B V2.0	4K / 21.6 GB	5	0
Hkcode Meta QA5	4K / 21.6 GB	10	0
Eeve Leaderboard Inst V1.5	4K / 21.6 GB	2157	0
...struct 10.8B V1.0 Empathy V0.1	4K / 21.6 GB	7	0
EEVE 10.8 BOOK V0.1	4K / 21.6 GB	7	1
AlgograpV4	4K / 43.2 GB	8	0
EEVE Instruct Math 10.8B	4K / 21.6 GB	9	3

Note: green Score (e.g. "73.2") means that the model is better than maywell/EEVE-Korean-Instruct-10.8B-Var.