PEFT
7.7🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.
Fine-tuning and training tools, each scored against our published rubric. Our fine-tuning guide prices one 30-million-token run. It sets managed platforms against a rented GPU hour.
15 tools, each scored against our published methodologyThere are 15 Model Training tools here. 80% have a free tier, and the median score is 6.7. Open source is the most common pricing model at 73.3%.
The new AI tools and GitHub repos we scored, each with its score and a dated source. Free. We score against a published rubric and show the working. Tools we earn a commission on are labelled Our pick, and their scores never move for money.
You confirm by email, and every issue has a one-click unsubscribe. We never sell your address.
Ranked by AIRR score. Some links are affiliate links, which never affect the score.