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

Pygmalion 6B by PygmalionAI

 »  All LLMs  »  PygmalionAI  »  Pygmalion 6B   URL Share it on

  3075   

Pygmalion 6B is an open-source language model by PygmalionAI. Features: 6b LLM, VRAM: 16.3GB, License: creativeml-openrail-m, LLM Explorer Score: 0.15, Arc: 40.5, HellaSwag: 67.5, MMLU: 25.7, GSM8K: 2.1.

  Conversational   En   Gptj   Pytorch   Region:us   Sharded   Tensorboard
Model Card on HF 🤗: https://huggingface.co/PygmalionAI/pygmalion-6b 

Pygmalion 6B Benchmarks

ARC: 40.53 vs 96.7 (so35)-58.1%
HellaSwag: 67.47 vs 95.3 (gpt4)-29.2%
MMLU: 25.73 vs 88.3 (so35)-70.9%
GSM8K: 2.05 vs 96.4 (so35)-97.9%
LLME Score: 0.1488
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Pygmalion 6B (PygmalionAI/pygmalion-6b)
🌟 Advertise your project 🚀

Pygmalion 6B Parameters and Internals

Model Type 
text generation
Use Cases 
Primary Use Cases:
dialogue model, requires specific input formatting
Limitations:
This model is NOT suitable for use by minors. It **will** output X-rated content under certain circumstances.
Additional Notes 
Proof-of-concept dialogue model; known for outputting X-rated content under specific circumstances.
Supported Languages 
en (English)
Training Details 
Data Sources:
the uft-6b ConvoGPT model, 56MB of dialogue data from multiple sources
Data Volume:
~48.5 million tokens
Methodology:
fine-tuning
Hardware Used:
4 NVIDIA A40s using DeepSpeed
Input Output 
Input Format:
[CHARACTER]'s Persona: [A few sentences about the character you want the model to play] [DIALOGUE HISTORY] You: [Your input message here] [CHARACTER]:
Accepted Modalities:
text
Output Format:
text
Performance Tips:
The model performs best if input prompts adhere to a specific format, including persona and dialogue history.
LLM NamePygmalion 6B
Repository 🤗https://huggingface.co/PygmalionAI/pygmalion-6b 
Model Size6b
Required VRAM16.3 GB
Updated2026-04-29
MaintainerPygmalionAI
Model Typegptj
Model Files  10.4 GB: 1-of-2   5.9 GB: 2-of-2
Supported Languagesen
Model ArchitectureGPTJForCausalLM
Licensecreativeml-openrail-m
Model Max Length1024
Transformers Version4.26.0.dev0
Tokenizer ClassGPT2Tokenizer
Beginning of Sentence Token<|endoftext|>
End of Sentence Token<|endoftext|>
Unk Token<|endoftext|>
Vocabulary Size50400
Torch Data Typefloat16
Activation Functiongelu_new
Errorsreplace

Best Alternatives to Pygmalion 6B

Best Alternatives
Context / RAM
Downloads
Likes
GPT J 6B0K / 24.2 GB3058871524
Mlperf GPT J 6B0K / 24.1 GB115950
Deception Normal0K / 12.2 GB60
Deception Filteredpositive0K / 12.2 GB60
Gptj Allenai Toxicity Blackbox0K / 12.2 GB90
Pygmalion 6B Roleplay0K / 12.1 GB10042
Gpt4all J0K / 12.2 GB2343300
...j Allenai Toxicity Explainable0K / 12.2 GB50
Test GPT J 6B0K / 2.5 GB80
Dolly Shygmalion 6B Dev V8P20K / 12.1 GB18756
Note: green Score (e.g. "73.2") means that the model is better than PygmalionAI/pygmalion-6b.

Rank the Pygmalion 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? 53972 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.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a
  Support LLM Explorer