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

Em German Mistral V01 GPTQ by TheBloke

 »  All LLMs  »  TheBloke  »  Em German Mistral V01 GPTQ   URL Share it on

  9   
  4-bit   Autotrain compatible Base model:jphme/em german mis... Base model:quantized:jphme/em ...   De   Deutsch   German   Gptq   Llama   Llama2   Mistral   Pytorch   Quantized   Region:us   Safetensors
Model Card on HF 🤗: https://huggingface.co/TheBloke/em_german_mistral_v01-GPTQ 

Em German Mistral V01 GPTQ Benchmarks

LLME Score: 0.11539
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Em German Mistral V01 GPTQ (TheBloke/em_german_mistral_v01-GPTQ)
🌟 Advertise your project 🚀

Em German Mistral V01 GPTQ Parameters and Internals

Model Type 
text-generation
Use Cases 
Areas:
research, language understanding, language generation
Applications:
text generation for German language, chatbot and assistant applications
Primary Use Cases:
generating and understanding German content
Limitations:
may produce factually incorrect or biased outputs
Considerations:
The model excels at German text but has limitations inherent in current generation models.
Additional Notes 
This is an experimental version; features like GQA and sliding windows are missing. Use at your own discretion.
Supported Languages 
de (native)
Training Details 
Data Sources:
mixed dataset of German instructions, chat data, and synthetic data
Data Volume:
>3 billion tokens
Methodology:
fine-tuned on large German dataset
Context Length:
8192
Model Architecture:
Llama2 with Mistral basis, lacking GQA and sliding windows features
Responsible Ai Considerations 
Fairness:
The model may generate biased content; users should take care when using model outputs.
Accountability:
Model output accuracy and reliability are not guaranteed; users must ensure appropriateness of outputs for their applications.
Input Output 
Input Format:
Follows Vicuna format without line breaks for prompts. A special format is used for RAG.
Accepted Modalities:
text
Output Format:
Text response formatted for German language tasks, with specific prompts optimizing interaction.
Release Notes 
Version:
v01 (Alpha)
Notes:
Initial release of the German language model. Optimized for text generation in German. Fine-tuned on a large dataset of German instructions.
LLM NameEm German Mistral V01 GPTQ
Repository 🤗https://huggingface.co/TheBloke/em_german_mistral_v01-GPTQ 
Model NameEM German Mistral v01
Model CreatorJan Philipp Harries
Base Model(s)  Em German Mistral V01   jphme/em_german_mistral_v01
Model Size1.2b
Required VRAM4.2 GB
Updated2025-08-18
MaintainerTheBloke
Model Typemistral
Model Files  4.2 GB
Supported Languagesde
GPTQ QuantizationYes
Quantization Typegptq
Model ArchitectureMistralForCausalLM
Licenseapache-2.0
Context Length32768
Model Max Length32768
Transformers Version4.34.0
Tokenizer ClassLlamaTokenizer
Vocabulary Size32000
Torch Data Typefloat16

Best Alternatives to Em German Mistral V01 GPTQ

Best Alternatives
Context / RAM
Downloads
Likes
... Finetune 16bit Ver9 Main GPTQ32K / 4.2 GB50
Dictalm2.0 GPTQ32K / 4.2 GB2230
Dictalm2.0 Instruct GPTQ32K / 4.2 GB910
Multi Verse Model GPTQ32K / 4.2 GB31
Turdus GPTQ32K / 4.2 GB115
Garrulus GPTQ32K / 4.2 GB183
HamSter 0.1 GPTQ32K / 4.2 GB102
Phoenix GPTQ32K / 4.2 GB51
...hat 3.5 1210 Seraph Slerp GPTQ32K / 4.2 GB102
Mistral Ft Optimized 1227 GPTQ32K / 4.2 GB102
Note: green Score (e.g. "73.2") means that the model is better than TheBloke/em_german_mistral_v01-GPTQ.

Rank the Em German Mistral V01 GPTQ 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? 50728 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