Shark-NLP/OpenICL
An open-source framework for in-context learning research and prototyping with language models.
Not currently ranked — collecting fresh signals.
star history
OpenICL provides a unified interface for in-context learning experiments, integrating various retrieval and inference methods to enable systematic comparison of language models. It supports various prompt instructions, retrieval techniques, and inference strategies including self-consistency. Users can load datasets, define prompt templates, and evaluate different LM configurations with built-in methods for fast research prototyping.
Frequently asked
- What is Shark-NLP/OpenICL?
- An open-source framework for in-context learning research and prototyping with language models.
- Is OpenICL open source?
- Yes — Shark-NLP/OpenICL is open source, released under the Apache-2.0 license.
- What language is OpenICL written in?
- Shark-NLP/OpenICL is primarily written in Python.
- How popular is OpenICL?
- Shark-NLP/OpenICL has 589 stars on GitHub.
- Where can I find OpenICL?
- Shark-NLP/OpenICL is on GitHub at https://github.com/Shark-NLP/OpenICL.