metadata
language:
- en
tags:
- markdown
- information
- index
Model Catalogue
This repository contains a collection of boutique AI models and is organized as follows:
Pretrained Base Models for Text Embedding
English Models
These models were finetuned on a subset of Zyphra/Zyda-2:
Multilingual Models
These models were aligned using agentlans/en-translations:
Text Statistics Models
These models take text as input and output a number.
| Base Model | Language | Quality | Readability | Sentiment |
|---|---|---|---|---|
| deberta-v3-xsmall-zyda-2 | English | Link | Link | Link |
| deberta-v3-base-zyda-2 | English | Link | Link | Link |
| multilingual-e5-small-aligned | Multilingual | Link | Link | Link |
| mdeberta-v3-base | Multilingual | Link | Link | Link |
Note: The mdeberta-v3-base models were trained on a previous version of the dataset, not the complete dataset.
Small Text-to-Text Models (English Only)
These models take text as input and produce text as output.
| Task | Model | Dataset |
|---|---|---|
| Keyword extraction | flan-t5-small-keywords | wikipedia-paragraph-keywords |
| Title generation | flan-t5-small-title | wikipedia-paragraph-titles |
| Summarization | text-summarization | wikipedia-paragraph-summaries |
Natural Language Inference (NLI) Models (English Only)
These models take text as input and output a label (entailment, neutral, or contradiction).