README.md

metadata

library_name: recurrentgemma
extra_gated_heading: Access RecurrentGemma on Hugging Face
extra_gated_prompt: >-
  To access RecurrentGemma on Hugging Face, you’re required to review and agree
  to Google’s usage license. To do this, please ensure you’re logged-in to
  Hugging Face and click below. Requests are processed immediately.
extra_gated_button_content: Acknowledge license
license: gemma
license_link: https://ai.google.dev/gemma/terms
pipeline_tag: text-generation
tags:
  - jax
  - recurrent_gemma
  - conversational

RecurrentGemma Model Card

Model Page: RecurrentGemma

This repository corresponds to the research RecurrentGemma repository in Jax.

This model card corresponds to the 2B instruct version of the RecurrentGemma model for usage with flax. For more information about the model, visit https://huggingface.co/google/recurrentgemma-2b-it.

Resources and Technical Documentation:

• Responsible Generative AI Toolkit
• Gemma on Kaggle
• Gemma on Vertex Model Garden

Terms of Use: Terms

Authors: Google

Loading the model

To download the weights and tokenizer, run:

from huggingface_hub import snapshot_download

local_dir = snapshot_download(repo_id="google/recurrentgemma-2b-it-flax")
snapshot_download(repo_id="google/recurrentgemma-2b-it-flax", local_dir=local_dir)

Then download this script from the gemma GitHub repository and call python sampling.py with the --path_checkpoint and --path_tokenizer arguments pointing to your local download path.