ResPulse Offline AI

Mobile-ready model assets for ResPulse's optional offline assistant and local RAG.

Files

  • Gemma3-1B-IT_multi-prefill-seq_q4_ekv4096.litertlm โ€” Gemma 3 1B instruction-tuned LiteRT-LM generator (Q4, EKV 4096).
  • embeddinggemma-300M_seq1024_mixed-precision.tflite โ€” EmbeddingGemma 300M LiteRT embedder.
  • sentencepiece.model โ€” tokenizer used by the embedding model.

Source and attribution

These files are unmodified copies redistributed from:

Gemma was created by Google. The LiteRT conversions were published by the LiteRT Community. This repository is a convenience mirror for the ResPulse mobile app and is not endorsed by Google or Hugging Face.

Terms and use restrictions

Downloading or using these files means you agree to the Gemma Terms of Use included in GEMMA_TERMS_OF_USE.md and published at https://ai.google.dev/gemma/terms. Use is also subject to the Gemma Prohibited Use Policy at https://ai.google.dev/gemma/prohibited_use_policy and applicable law.

The model can produce inaccurate output. Do not treat it as professional medical, legal, emergency, or safety advice.

Downloads last month
21
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support