README.md

metadata

library_name: transformers
base_model: p-e-w/gemma-4-E2B-it-heretic-ara
tags:
  - onnx
  - webgpu
  - transformers
  - transformers.js
  - quantized
  - text
  - image
  - gemma4

rally-2b

Browser-oriented ONNX export of a Gemma 4 Heretic checkpoint packaged for WebGPU / Transformers.js.

Capabilities

• Supported inputs: text, image
• Not included in this package: audio, video

Version Notes

• This is the stable v1 browser package.
• For the enhanced multimodal variant, see thomasjvu/rally-2b-v2.
• The v2 package adds audio, video for this model family.

Provenance

• Source model: p-e-w/gemma-4-E2B-it-heretic-ara
• Base model for inherited processor assets: google/gemma-4-E2B-it
• Architecture family: gemma4_conditional_generation
• Expected architecture: Gemma4ForConditionalGeneration
• Target dtype: q4f16
• Target device: webgpu

Expected ONNX Sessions

• vision_encoder_q4f16.onnx
• embed_tokens_q4f16.onnx
• decoder_model_merged_q4f16.onnx

Usage

Load this repo with Transformers.js using the model's transformers.js_config metadata and WebGPU backend.