Thomas van Dongen
  • Projects
  • Blog
  • About

Projects

Open-source packages I build and maintain, mostly under Minish.

  • Semble Code search for agents that uses ~99% fewer tokens than grep+read.
    6.2k 549.8k
  • Model2Vec Distil any sentence transformer into a tiny, fast static embedding model.
    2.2k 6.9M
  • SemHash Semantic deduplication and dataset filtering across text, images, and audio.
    972 873.3k
  • Pyversity Diversify search and retrieval results to reduce redundancy and improve coverage.
    501 537.6k
  • Vicinity Fast, lightweight nearest neighbor search with pluggable backends.
    351 1.2M
  • Model2Vec-rs A Rust port of Model2Vec.
    217 7.3k
  • Tokenlearn Pre-train static embedding models for distillation pipelines.
    111 8.2k
  • Agentcheck Scans your shell and reports what an AI agent could access, by severity.
    28 2.3k
  • Potion Tiny state-of-the-art static embedding models for English, multilingual, code, and retrieval.
© 2026 Thomas van Dongen