What are the hardware requirements for Linkbricks Horizon AI Korean Advanced 27B?

Linkbricks Horizon AI Korean Advanced 27B requires approximately 54.7 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 Linkbricks Horizon AI Korean Advanced 27B and how large is it?

Linkbricks Horizon AI Korean Advanced 27B is developed by Saxo, a model with 27b 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 Linkbricks Horizon AI Korean Advanced 27B?

Linkbricks Horizon AI Korean Advanced 27B is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Linkbricks Horizon AI Korean Advanced 27B by Saxo — VRAM 54.7GB, 8K context

Name: Linkbricks Horizon AI Korean Advanced 27B
Author: Saxo

Linkbricks Horizon AI Korean Advanced 27B is an open-source language model by Saxo. Features: 27b LLM, VRAM: 54.7GB, Context: 8K, License: apache-2.0, Quantized, LLM Explorer Score: 0.14.

Model Card on HF 🤗: https://huggingface.co/Saxo/Linkbricks-Horizon-AI-Korean-Advanced-27B

Linkbricks Horizon AI Korean Advanced 27B Parameters and Internals

Model Type

text generation

Use Cases

Areas:

Research, Commercial applications

Applications:

Language translation, Mathematics, Logic judgment, Social media analysis, AI enhanced writing

Primary Use Cases:

Cross-language Processing, Complex problem-solving, High-dimensional analysis

Considerations:

Tokenizer does not expand vocabulary of the base model

Additional Notes

Tokenization remains the same as the base model without word expansion.

Supported Languages

ko (Korean), en (English), jp (Japanese), cn (Chinese)

Training Details

Data Sources:

Saxo/ko_cn_translation_tech_social_science_linkbricks_single_dataset, Saxo/ko_jp_translation_tech_social_science_linkbricks_single_dataset, Saxo/en_ko_translation_tech_science_linkbricks_single_dataset_with_prompt_text_huggingface, Saxo/en_ko_translation_social_science_linkbricks_single_dataset_with_prompt_text_huggingface, Saxo/ko_aspect_sentiment_sns_mall_sentiment_linkbricks_single_dataset_with_prompt_text_huggingface, Saxo/ko_summarization_linkbricks_single_dataset_with_prompt_text_huggingface, Saxo/OpenOrca_cleaned_kor_linkbricks_single_dataset_with_prompt_text_huggingface, Saxo/ko_government_qa_total_linkbricks_single_dataset_with_prompt_text_huggingface_sampled, Saxo/ko-news-corpus-1, Saxo/ko-news-corpus-2, Saxo/ko-news-corpus-3, Saxo/ko-news-corpus-4, Saxo/ko-news-corpus-5, Saxo/ko-news-corpus-6, Saxo/ko-news-corpus-7, Saxo/ko-news-corpus-8, Saxo/ko-news-corpus-9, maywell/ko_Ultrafeedback_binarized, youjunhyeok/ko-orca-pair-and-ultrafeedback-dpo, lilacai/glaive-function-calling-v2-sharegpt, kuotient/gsm8k-ko

Data Volume:

10M Korean news corpus

Methodology:

Continued-Pretraining (CPT), Supervised Fine-Tuning (SFT), Decision Point Optimization (DPO)

Context Length:

128000

Hardware Used:

8x H100-80G GPUs

Model Architecture:

gemma-2-27b-it based architecture with specialized cross-language training techniques

LLM Name	Linkbricks Horizon AI Korean Advanced 27B
Repository 🤗	https://huggingface.co/Saxo/Linkbricks-Horizon-AI-Korean-Advanced-27B
Base Model(s)	Gemma 2 27B It google/gemma-2-27b-it
Model Size	27b
Required VRAM	54.7 GB
Updated	2026-08-02
Maintainer	Saxo
Model Type	gemma2
Model Files	16.6 GB 19.4 GB 22.3 GB 28.9 GB 4.7 GB: 1-of-12 4.9 GB: 2-of-12 4.9 GB: 3-of-12 5.0 GB: 4-of-12 4.9 GB: 5-of-12 4.9 GB: 6-of-12 5.0 GB: 7-of-12 4.9 GB: 8-of-12 4.9 GB: 9-of-12 5.0 GB: 10-of-12 4.9 GB: 11-of-12 0.7 GB: 12-of-12
Supported Languages	ko en jp cn
GGUF Quantization	Yes
Quantization Type	gguf
Model Architecture	Gemma2ForCausalLM
License	apache-2.0
Context Length	8192
Model Max Length	8192
Transformers Version	4.43.2
Tokenizer Class	GemmaTokenizer
Padding Token	<pad>
Vocabulary Size	256000
Torch Data Type	bfloat16

Rank the Linkbricks Horizon AI Korean Advanced 27B 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 Linkbricks Horizon AI Korean Advanced 27B

Best Alternatives	Context / RAM	Downloads	Likes
Gemma 2 27B It GGUF	8K / 10.4 GB	327	7
Gemma 2 27B It GGUF	8K / 10.4 GB	249	0
Gemma 2 27B 8bit	8K / 28.8 GB	11	2
Gemma 2 27B 4bit	8K / 15.3 GB	9	0
Gemma2 Mixed Therapy Ft	8K / 15.8 GB	5	2
Gemma 2 27B It Bnb 4bit	8K / 15.8 GB	830	11
GEMMA2 27B NLI 16bit	8K / 54.7 GB	11	0
Gemma 2 27B It 4bit	8K / 15.3 GB	372	8
Gemma 2 27B It 8bit	8K / 28.8 GB	152	10
Gemma 2 27B Bnb 4bit	8K / 15.8 GB	90	13

Note: green Score (e.g. "73.2") means that the model is better than Saxo/Linkbricks-Horizon-AI-Korean-Advanced-27B.