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

DiffMerge DollyGPT Pygmalion by TehVenom

 »  All LLMs  »  TehVenom  »  DiffMerge DollyGPT Pygmalion   URL Share it on

  853   

DiffMerge DollyGPT Pygmalion is an open-source language model by TehVenom. Features: LLM, VRAM: 12.1GB, HF Score: 30.5, LLM Explorer Score: 0.12, Arc: 23.6, HellaSwag: 34.4, MMLU: 24.4, TruthfulQA: 46.5, WinoGrande: 53.8.

  Endpoints compatible   Gptj   Pytorch   Region:us   Sharded
Model Card on HF 🤗: https://huggingface.co/TehVenom/DiffMerge-DollyGPT-Pygmalion 

DiffMerge DollyGPT Pygmalion Benchmarks

ARC: 23.63 vs 96.7 (so35)-75.6%
HellaSwag: 34.38 vs 95.3 (gpt4)-63.9%
MMLU: 24.41 vs 88.3 (so35)-72.4%
TruthfulQA: 46.48 vs 59 (gpt4)-21.2%
WinoGrande: 53.83 vs 87.5 (gpt4)-38.5%
LLME Score: 0.12114
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
DiffMerge DollyGPT Pygmalion (TehVenom/DiffMerge-DollyGPT-Pygmalion)
🌟 Advertise your project 🚀

DiffMerge DollyGPT Pygmalion Parameters and Internals

Model Type 
Text Generation
Use Cases 
Areas:
Research, Development
Additional Notes 
This model is a theoretical construction based on a weighted diff merge of existing language models. Practical implementation would require advanced knowledge in ML and resources.
Supported Languages 
English (Primary)
Training Details 
Data Sources:
Combination of data sources used in GPT-J-Dolly and Pygmalion-6b
Methodology:
Diff merge technique applied: ((GPT-J-Dolly - GPT-J-6b)*0.7 + Pygmalion-6b*0.9) - Pygmalion-6b
Hardware Used:
GPUs compatible for 6 billion parameter models
Model Architecture:
Derived from architecture of GPT-J-6b.
LLM NameDiffMerge DollyGPT Pygmalion
Repository 🤗https://huggingface.co/TehVenom/DiffMerge-DollyGPT-Pygmalion 
Required VRAM12.1 GB
Updated2026-03-23
MaintainerTehVenom
Model Typegptj
Model Files  2.1 GB: 1-of-6   2.1 GB: 2-of-6   2.0 GB: 3-of-6   2.0 GB: 4-of-6   2.0 GB: 5-of-6   1.9 GB: 6-of-6
Model ArchitectureGPTJForCausalLM
Model Max Length1024
Transformers Version4.28.0.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 DiffMerge DollyGPT Pygmalion

Best Alternatives
Context / RAM
Downloads
Likes
...Merge Pygmalion Main Onto V8P40K / 12.3 GB8691
ChanMalion0K / 12.3 GB10359
Skegma GPTJ0K / 12.3 GB8430
Javalion R0K / 12.3 GB8585
Adventien GPTJ0K / 12.3 GB9000
Javalion GPTJ0K / 12.3 GB8261
Javelin R0K / 12.3 GB8952
Janin GPTJ0K / 12.3 GB8530
Javelin GPTJ0K / 12.3 GB8294
Janin R0K / 12.3 GB8401

Rank the DiffMerge DollyGPT Pygmalion 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? 52473 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