What are the hardware requirements for Gemma Ko Merge?

Gemma Ko Merge requires approximately 20.4 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 Gemma Ko Merge and how large is it?

Gemma Ko Merge is developed by Gunulhona, a model with 9b 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 Gemma Ko Merge?

Gemma Ko Merge is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Gemma Ko Merge by Gunulhona — VRAM 20.4GB, 8K context

Name: Gemma Ko Merge
Author: Gunulhona

Gemma Ko Merge is an open-source language model by Gunulhona. Features: 9b LLM, VRAM: 20.4GB, Context: 8K, Quantized, Merged, LLM Explorer Score: 0.22.

Merged Model Base model:lemon07r/gemma-2-at... Base model:rtzr/ko-gemma-2-9b-... Base model:wzhouad/gemma-2-9b-... Conversational Deploy:azure Endpoints compatible Gemma2 Gguf Model-index Quantized Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/Gunulhona/Gemma-Ko-Merge

Gemma Ko Merge Parameters and Internals

Additional Notes

This model is a merge of pre-trained language models created using mergekit. The merge was conducted using the breadcrumbs_ties merge method.

LLM Name	Gemma Ko Merge
Repository 🤗	https://huggingface.co/Gunulhona/Gemma-Ko-Merge
Base Model(s)	Gemma 2 Ataraxy 9B Gemma 2 9B It WPO HB Ko Gemma 2 9B It ...emma9 On Korean Summary Events Ko Gemma 2 9B It lemon07r/Gemma-2-Ataraxy-9B wzhouad/gemma-2-9b-it-WPO-HB rtzr/ko-gemma-2-9b-it ghost613/gemma9_on_korean_summary_events rtzr/ko-gemma-2-9b-it
Merged Model	Yes
Model Size	9b
Required VRAM	20.4 GB
Updated	2026-07-14
Maintainer	Gunulhona
Model Type	gemma2
Model Files	9.8 GB 1.8 GB: 1-of-11 1.9 GB: 2-of-11 2.0 GB: 3-of-11 2.0 GB: 4-of-11 2.0 GB: 5-of-11 2.0 GB: 6-of-11 2.0 GB: 7-of-11 2.0 GB: 8-of-11 2.0 GB: 9-of-11 2.0 GB: 10-of-11 0.7 GB: 11-of-11
GGUF Quantization	Yes
Quantization Type	gguf
Model Architecture	Gemma2ForCausalLM
Context Length	8192
Model Max Length	8192
Transformers Version	4.45.0.dev0
Tokenizer Class	GemmaTokenizer
Padding Token	<pad>
Vocabulary Size	256000
Torch Data Type	bfloat16

Rank the Gemma Ko Merge 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 Gemma Ko Merge

Best Alternatives	Context / RAM	Downloads	Likes
...LLM X Gemma 2 9B CyberSecurity	8K / 18.6 GB	50	1
... Russian Function Calling GGUF	8K / 18.6 GB	1760	30
Gemma 2 9B It GGUF	8K / 3.8 GB	528	4
Gemma 2 9B Chinese Chat GGUF	8K / 3.8 GB	253	3
Gemma 2 9B It GGUF	8K / 3.8 GB	294	0
Gemma 2 9B Chinese Chat GGUF	8K / 3.8 GB	152	0
GWQ 9B Preview	8K / 18.6 GB	9	3
Gemma 2 9B 8bit	8K / 9.8 GB	178	9
Gemma 2 9B 4bit	8K / 5.2 GB	160	2
Gemma 2 9B It Bnb 4bit	8K / 6.1 GB	13007	32

Note: green Score (e.g. "73.2") means that the model is better than Gunulhona/Gemma-Ko-Merge.