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

Unbabel TowerInstruct 7B V0.1 4bits by RichardErkhov

 »  All LLMs  »  RichardErkhov  »  Unbabel TowerInstruct 7B V0.1 4bits   URL Share it on

  81   
  Arxiv:2402.17733   4-bit   Autotrain compatible   Bitsandbytes   Conversational   Endpoints compatible   Llama   Region:us   Safetensors
Model Card on HF 🤗: https://huggingface.co/RichardErkhov/Unbabel_-_TowerInstruct-7B-v0.1-4bits 

Unbabel TowerInstruct 7B V0.1 4bits Benchmarks

LLME Score: 0.18759
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Unbabel TowerInstruct 7B V0.1 4bits (RichardErkhov/Unbabel_-_TowerInstruct-7B-v0.1-4bits)

Unbabel TowerInstruct 7B V0.1 4bits Parameters and Internals

Model Type 
Translation, Supervised fine-tuning
Use Cases 
Areas:
Research, Translation
Applications:
Machine Translation, Automatic Post Edition, Grammatical Error Correction, Paraphrase Generation, Named-entity Recognition
Primary Use Cases:
Sentence-level Translation, Paragraph-level Translation, Context-aware Translation, Terminology-aware Translation
Limitations:
Not aligned to human preferences, may generate problematic outputs., Not intended for document-level translation
Additional Notes 
Initially fine-tuned for translation-related tasks and augmented with conversational datasets and code instructions.
Supported Languages 
English (High), Portuguese (High), Spanish (High), French (High), German (High), Dutch (High), Italian (High), Korean (High), Chinese (High), Russian (High)
Training Details 
Data Sources:
TowerBlocks Dataset
Methodology:
Fine-tuning TowerBase on publicly available and synthetic datasets specific to translation-related tasks.
Context Length:
2048
Input Output 
Input Format:
ChatML prompt templates without system prompts.
Output Format:
Generates translated text according to the user prompt.
LLM NameUnbabel TowerInstruct 7B V0.1 4bits
Repository 🤗https://huggingface.co/RichardErkhov/Unbabel_-_TowerInstruct-7B-v0.1-4bits 
Model Size7b
Required VRAM4.2 GB
Updated2025-02-22
MaintainerRichardErkhov
Model Typellama
Model Files  4.2 GB
Supported Languagesen de fr zh pt nl ru ko it es
Model ArchitectureLlamaForCausalLM
Licensecc-by-nc-4.0
Context Length4096
Model Max Length4096
Transformers Version4.40.2
Tokenizer ClassLlamaTokenizer
Padding Token<|im_end|>
Vocabulary Size32007
Torch Data Typefloat16

Best Alternatives to Unbabel TowerInstruct 7B V0.1 4bits

Best Alternatives
Context / RAM
Downloads
Likes
2 Very Sci Fi1024K / 16.1 GB3170
...1M 1000000ctx AEZAKMI 3 1 17021024K / 13.5 GB231
... Qwen2.5llamaify 7B V23.1 200K195K / 15.2 GB39433
LlamaStock 8B128K / 16.1 GB111
SuperNeuralDreadDevil 8B128K / 16.1 GB541
Yarn Llama 2 7B 128K128K / 13.5 GB642239
LLaMA 7B PoSE YaRN 128K128K / 13.5 GB73
LLaMA 7B PoSE Linear 96K96K / 27 GB92
LLaMA 7B PoSE YaRN 96K96K / 13.5 GB111
Chat Llama2 7B 80K80K / 13.8 GB80
Note: green Score (e.g. "73.2") means that the model is better than RichardErkhov/Unbabel_-_TowerInstruct-7B-v0.1-4bits.

Rank the Unbabel TowerInstruct 7B V0.1 4bits 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