minimaxir/gpt-2-simple

Health v2 · maintenance only

Flags: no_license

Adoption not part of the score

3400 stars · 666 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

A Python package that simplifies fine-tuning OpenAI's GPT-2 text-generation model (124M/355M) on custom text and generating text from the retrained model, built on TensorFlow. Development has largely been superseded by the same author's aitextgen.

Use cases

• fine-tune gpt-2 on my own text corpus
• generate text with a retrained gpt-2 model
• train a small language model on a custom dataset
• generate text continuations from a prompt prefix
• run gpt-2 training in a colab notebook

When to choose

• you specifically need TensorFlow-based GPT-2 fine-tuning
• you want a simple, scriptable API for retraining and generating with GPT-2
• you have existing gpt-2-simple checkpoints to continue using

When to avoid

• you want modern, efficient LLM fine-tuning - use aitextgen or Hugging Face transformers instead
• you need models larger than GPT-2 medium (355M)
• you want actively developed features or PyTorch support

Facets

library · maturity maintenance

llm-training llm-inference machine-learning deep-learning large-language-models deep-learning machine-learning python cross-platform gpt-2 fine-tuning text-generation tensorflow superseded-by-aitextgen natural-language-processing gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="minimaxir/gpt-2-simple")