EleutherAI/gpt-neox

An implementation of model parallel autoregressive transformers on GPUs, based on the Megatron and DeepSpeed libraries

58
/ 100
Established

This is a specialized toolkit for researchers and engineers who need to train very large language models from scratch, or fine-tune existing ones, using substantial computational resources. It takes raw text data and configuration settings as input, and outputs a custom-trained language model capable of generating human-like text. This is for users operating at the cutting edge of AI, often in academic, industry, or government labs.

7,399 stars.

Use this if you are a researcher or engineer looking to train large-scale language models with billions of parameters using distributed training across multiple GPUs or high-performance computing clusters.

Not ideal if you are looking to run generic inference with existing models or train smaller models; in those cases, simpler libraries like Hugging Face's `transformers` are more appropriate.

large-language-model-training deep-learning-research natural-language-processing high-performance-computing AI-model-development
No Package No Dependents
Maintenance 10 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 22 / 25

How are scores calculated?

Stars

7,399

Forks

1,100

Language

Python

License

Apache-2.0

Last pushed

Feb 03, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/EleutherAI/gpt-neox"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.