Pygmalion 6B Roleplay is an open-source language model by anhnv125. Features: 6b LLM, VRAM: 12.1GB, Merged, HF Score: 38.3, LLM Explorer Score: 0.13, Arc: 40.5, HellaSwag: 67.5, MMLU: 25.7, TruthfulQA: 32.5, WinoGrande: 62.7, GSM8K: 1.1.
Model Card on HF 🤗: https://huggingface.co/anhnv125/pygmalion-6b-roleplay
Pygmalion 6B Roleplay 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%
TruthfulQA: 32.53 vs 59 (gpt4)-44.9%
WinoGrande: 62.67 vs 87.5 (gpt4)-28.4%
GSM8K: 1.14 vs 96.4 (so35)-98.8%
LLME Score: 0.13147
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
What is the LLM Explorer Rank (Score)
Pygmalion 6B Roleplay Parameters and Internals
| LLM Name | Pygmalion 6B Roleplay |
| Repository 🤗 | https://huggingface.co/anhnv125/pygmalion-6b-roleplay |
| Merged Model | Yes |
| Model Size | 6b |
| Required VRAM | 12.1 GB |
| Updated | 2025-11-04 |
| Maintainer | anhnv125 |
| Model Type | gptj |
| Model Files | |
| Model Architecture | GPTJForCausalLM |
| Model Max Length | 1024 |
| Transformers Version | 4.32.0.dev0 |
| Tokenizer Class | GPT2Tokenizer |
| Beginning of Sentence Token | <|endoftext|> |
| End of Sentence Token | <|endoftext|> |
| Unk Token | <|endoftext|> |
| Vocabulary Size | 50432 |
| Torch Data Type | float16 |
| Activation Function | gelu_new |
| Errors | replace |
Best Alternatives to Pygmalion 6B Roleplay
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Mlperf GPT J 6B | 0K / 24.1 GB | 11595 | 0 |
| GPT J 6B | 0K / 24.2 GB | 43060 | 1509 |
| Pygmalion 6B | 0K / 16.3 GB | 914 | 752 |
| GPT JT 6B V1 | 0K / 12.2 GB | 9662 | 302 |
| Deception Filteredpositive | 0K / 12.2 GB | 9 | 0 |
| Deception Normal | 0K / 12.2 GB | 6 | 0 |
| Gptj Allenai Toxicity Blackbox | 0K / 12.2 GB | 9 | 0 |
| Gpt4all J | 0K / 12.2 GB | 2343 | 300 |
| ...j Allenai Toxicity Explainable | 0K / 12.2 GB | 7 | 0 |
| Pyg Instruct Wizardlm | 0K / 12.2 GB | 1834 | 1 |
Note: green Score (e.g. "73.2") means that the model is better than anhnv125/pygmalion-6b-roleplay .
Rank the Pygmalion 6B Roleplay 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! 🌟
What open-source LLMs or SLMs are you in search of? 52473 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a