OpenNLG/OpenBA-v2
OpenBA-V2: 3B LLM (Large Language Model) with T5 architecture, utilizing model pruning technique and continuing pretraining from OpenBA-15B.
This project helps machine learning engineers and researchers manage large language models more efficiently. It takes an existing, larger model like OpenBA-15B and uses pruning techniques to create a much smaller, 3.4B parameter version, OpenBA-v2. The output is a highly compressed, yet performant, large language model that is easier to deploy and run, especially for tasks like named entity recognition and question answering.
No commits in the last 6 months.
Use this if you need to run large language models for tasks like text generation or information extraction but are constrained by computational resources or model size.
Not ideal if you require the absolute bleeding-edge performance of much larger, un-pruned models, as there is a slight trade-off in certain benchmarks.
Stars
25
Forks
1
Language
Python
License
Apache-2.0
Category
Last pushed
May 10, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/OpenNLG/OpenBA-v2"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
ModelTC/LightCompress
[EMNLP 2024 & AAAI 2026] A powerful toolkit for compressing large models including LLMs, VLMs,...
p-e-w/heretic
Fully automatic censorship removal for language models
Orion-zhen/abliteration
Make abliterated models with transformers, easy and fast
YerbaPage/LongCodeZip
LongCodeZip: Compress Long Context for Code Language Models [ASE2025]
locuslab/wanda
A simple and effective LLM pruning approach.