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

Llama3 DiscoLeo Instruct 8B 32K V0.1 4bit AWQ by DiscoResearch

 »  All LLMs  »  DiscoResearch  »  Llama3 DiscoLeo Instruct 8B 32K V0.1 4bit AWQ   URL Share it on

  8   
  4-bit   4bit   Autotrain compatible   Awq   Conversational   De   En   Endpoints compatible   Instruct   Llama   Quantized   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/DiscoResearch/Llama3-DiscoLeo-Instruct-8B-32k-v0.1-4bit-awq 

Llama3 DiscoLeo Instruct 8B 32K V0.1 4bit AWQ Benchmarks

LLME Score: 0.15611
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Llama3 DiscoLeo Instruct 8B 32K V0.1 4bit AWQ (DiscoResearch/Llama3-DiscoLeo-Instruct-8B-32k-v0.1-4bit-awq)
🌟 Advertise your project 🚀

Llama3 DiscoLeo Instruct 8B 32K V0.1 4bit AWQ Parameters and Internals

Model Type 
Instruction-tuned
Use Cases 
Areas:
Research, Commercial applications
Applications:
German text generation, Bilingual tasks
Primary Use Cases:
German economic reports, Educational content creation
Additional Notes 
The model is a 4-bit quantization and is optimized for bilingual tasks.
Supported Languages 
de (High), en (Medium)
Training Details 
Data Sources:
DiscoLM German dataset, Occiglot fineweb dataset
Methodology:
Instruction tuning
Context Length:
8192
Hardware Used:
hessian.Ai 42 Supercomputer
Input Output 
Input Format:
Llama-3 chat template
Accepted Modalities:
text
Output Format:
text
LLM NameLlama3 DiscoLeo Instruct 8B 32K V0.1 4bit AWQ
Repository 🤗https://huggingface.co/DiscoResearch/Llama3-DiscoLeo-Instruct-8B-32k-v0.1-4bit-awq 
Base Model(s)  DiscoResearch/Llama3-DiscoLeo-Instruct-8B-32k-v0.1   DiscoResearch/Llama3-DiscoLeo-Instruct-8B-32k-v0.1
Model Size8b
Required VRAM5.8 GB
Updated2025-09-23
MaintainerDiscoResearch
Model Typellama
Instruction-BasedYes
Model Files  4.7 GB: 1-of-2   1.1 GB: 2-of-2
Supported Languagesde en
AWQ QuantizationYes
Quantization Typeawq|4bit
Model ArchitectureLlamaForCausalLM
Licensellama3
Context Length65536
Model Max Length65536
Transformers Version4.40.2
Tokenizer ClassPreTrainedTokenizerFast
Padding Token<|end_of_text|>
Vocabulary Size128256
Torch Data Typebfloat16

Best Alternatives to Llama3 DiscoLeo Instruct 8B 32K V0.1 4bit AWQ

Best Alternatives
Context / RAM
Downloads
Likes
...8B Instruct Gradient 1048K AWQ1024K / 5.8 GB70
...radient 1048K AWQ 4bit Smashed1024K / 5.8 GB71
...Instruct 262K AWQ 4bit Smashed256K / 5.8 GB54
...ta Llama 3 8B Instruct 64K AWQ64K / 5.8 GB60
Llama 3 8B Instruct AWQ8K / 5.8 GB1659825
...eta Llama 3 8B Instruct Hf AWQ8K / 5.8 GB14329
Meta Llama 3 8B Instruct AWQ8K / 5.8 GB12675
Meta Llama 3 8B Instruct AWQ8K / 5.8 GB60
Meta Llama 3 8B Instruct AWQ8K / 5.8 GB70
Meta Llama 3 8B Instruct AWQ8K / 5.8 GB2780
Note: green Score (e.g. "73.2") means that the model is better than DiscoResearch/Llama3-DiscoLeo-Instruct-8B-32k-v0.1-4bit-awq.

Rank the Llama3 DiscoLeo Instruct 8B 32K V0.1 4bit AWQ 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? 51544 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 v20241124
  Support LLM Explorer