LLM EXPLORER
 LLM Explorer: A Curated Large Language Model Directory and Analytics
Dark Theme
  LLM  List    LLM Hosting    LLM Leaderboards    Blog    Newsfeed    Advertise

Yi Ko 6B by beomi

 »  All LLMs  »  beomi  »  Yi Ko 6B   URL Share it on

  5203   
  01-ai   Autotrain compatible   Doi:10.57967/hf/1708   En   Ko   Llama   Model-index   Pytorch   Region:us   Safetensors   Sharded   Tensorflow   Yi   Yi-ko
Model Card on HF 🤗: https://huggingface.co/beomi/Yi-Ko-6B 

Yi Ko 6B Benchmarks

ARC: 48.89 vs 96.7 (so35)-49.4%
HellaSwag: 74.48 vs 95.3 (gpt4)-21.8%
MMLU: 55.72 vs 88.3 (so35)-36.9%
TruthfulQA: 37.09 vs 59 (gpt4)-37.1%
WinoGrande: 72.93 vs 87.5 (gpt4)-16.7%
GSM8K: 12.51 vs 96.4 (so35)-87%
LLME Score: 0.21302
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Yi Ko 6B (beomi/Yi-Ko-6B)

Yi Ko 6B Parameters and Internals

Model Type 
text-generation
Use Cases 
Areas:
Research
Primary Use Cases:
Text generation
Additional Notes 
Yi-Ko series models are advanced iterations of 01-ai/Yi models with expanded vocabulary including Korean/English corpus.
Supported Languages 
en (English), ko (Korean)
Training Details 
Data Sources:
Korean + English online data
Data Volume:
>60B tokens
Methodology:
Auto-regressive language model based on optimized transformer architecture similar to Llama2
Context Length:
4000
Model Architecture:
Transformer architecture based on Llama2, utilized via LlamaForCausalLM
Input Output 
Input Format:
Text only
Accepted Modalities:
text
Output Format:
Text generation
Release Notes 
Version:
Revision 205083a
Date:
2024-01-29
Notes:
New Model beomi/Yi-Ko-DUS-9B Released!
Version:
Revision 205083a
Date:
2023-12-03
Notes:
Yi-Ko(KoEN)-6B Achieved #1 Pretrained Models at Open Korean LLM Leaderboard!
Version:
Revision 205083a
Date:
2023-12-01
Notes:
Alpha Release of Yi-Ko(KoEN)-6B model
LLM NameYi Ko 6B
Repository 🤗https://huggingface.co/beomi/Yi-Ko-6B 
Model Size6b
Required VRAM12.4 GB
Updated2025-02-22
Maintainerbeomi
Model Typellama
Model Files  3.0 GB: 1-of-5   2.9 GB: 2-of-5   3.0 GB: 3-of-5   2.9 GB: 4-of-5   0.6 GB: 5-of-5   3.0 GB: 1-of-5   2.9 GB: 2-of-5   3.0 GB: 3-of-5   2.9 GB: 4-of-5   0.6 GB: 5-of-5
Supported Languagesen ko
Model ArchitectureLlamaForCausalLM
Licenseapache-2.0
Context Length2048
Model Max Length2048
Transformers Version4.33.1
Tokenizer ClassLlamaTokenizer
Padding Token<unk>
Vocabulary Size78464
Torch Data Typebfloat16

Best Alternatives to Yi Ko 6B

Best Alternatives
Context / RAM
Downloads
Likes
Yi 6B 200K195K / 12.1 GB8575172
Yi 6B 200K AEZAKMI V2195K / 12.1 GB17591
Yi 6B 200K DPO195K / 12.1 GB19660
Wukong Yi 6B 200K195K / 12.1 GB211
Barcenas 6B 200K195K / 12.1 GB19382
Yi 6B 200K Llamafied195K / 12.1 GB2711
Yi 6B 200K Llama195K / 12.1 GB165
Llama 3.2 6B AlgoCode128K / 12.7 GB9868
Chatglm2 6B Port Llama32K / 12.5 GB154
Miqu 6B Truthy31K / 11.3 GB1011
Note: green Score (e.g. "73.2") means that the model is better than beomi/Yi-Ko-6B.

Rank the Yi Ko 6B Capabilities

🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟

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  

What open-source LLMs or SLMs are you in search of? 43470 in total.

Email us: info@extractum.io. Our Privacy Policy  | Terms and Conditions  | Suggest an improvement.
Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241227
  Support LLM Explorer