Dharmamitra releases a Qwen3.5 model family for Buddhist text translation and retrieval

SebastianNehrd2 · x · 2026-07-29

Dharmamitra released a new open-source Qwen3.5-based model family for classical Buddhist literature, including base, instruction, translation, and embedding models.

According to the post, the team used more than 30B tokens for continuous pretraining, and the collection includes 9B and 2B variants for translation, instruction following, and retrieval over Buddhist texts.

Original post →

More from Models

Models channel →