What are the hardware requirements for Pygmalion 350M?

Pygmalion 350M requires approximately 1.3 GB of VRAM and supports a context window of 2K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Pygmalion 350M and how large is it?

Pygmalion 350M is developed by PygmalionAI, a model with 350m parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does Pygmalion 350M perform on standard benchmarks?

Pygmalion 350M has the following published scores: MMLU 25.68. Compare against reference models on this page or on the LLM Explorer leaderboards.

Pygmalion 350M by PygmalionAI — VRAM 1.3GB, 2K context

Name: Pygmalion 350M
Author: PygmalionAI

Pygmalion 350M is an open-source language model by PygmalionAI. Features: 350m LLM, VRAM: 1.3GB, Context: 2K, LLM Explorer Score: 0.1, Arc: 25, HellaSwag: 37.8, MMLU: 25.7, GSM8K: 0.5.

Convai Conversational En Opt Pytorch Region:us

Model Card on HF 🤗: https://huggingface.co/PygmalionAI/pygmalion-350m

Pygmalion 350M Parameters and Internals

Model Type

conversational

Additional Notes

NSFW data was included in fine-tuning. SFW inputs may result in SFW outputs, but there's no guarantee.

Training Details

Data Sources:

dialogue datasets from various sources

Data Volume:

273 KB

Methodology:

Fine-tuning using ColossalAI library

Training Time:

less than an hour

Hardware Used:

singular GPU with 6 GB of VRAM

LLM Name	Pygmalion 350M
Repository 🤗	https://huggingface.co/PygmalionAI/pygmalion-350m
Model Size	350m
Required VRAM	1.3 GB
Updated	2026-07-28
Maintainer	PygmalionAI
Model Type	opt
Model Files	1.3 GB
Supported Languages	en
Model Architecture	OPTForCausalLM
Context Length	2048
Model Max Length	2048
Transformers Version	4.20.0.dev0
Beginning of Sentence Token	</s>
End of Sentence Token	</s>
Unk Token	</s>
Vocabulary Size	50272
Torch Data Type	float16
Activation Function	relu
Errors	replace

Rank the Pygmalion 350M capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

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

Best Alternatives to Pygmalion 350M

Best Alternatives	Context / RAM	Downloads	Likes
Opt 350M	2K / 0.7 GB	198521	149
Opt Mini Dataset 0	2K / 0.7 GB	8	0
Facebook Opt 350M SFT Korz14	2K / 0.7 GB	12	0
Temp Model Sft	2K / 1.3 GB	7	0
Gpt350 Chat S V0	2K / 0.7 GB	6	0
Gpt350 Chat S V0 1	2K / 0.7 GB	5	0
Dadjokes Tuned Opt	2K / 1.3 GB	8	2
Rockyalquimista888	2K / 1.3 GB	8	0
Remedycure	2K / 1.3 GB	6	0
Sunday	2K / 1.3 GB	6	0

Note: green Score (e.g. "73.2") means that the model is better than PygmalionAI/pygmalion-350m.