File size: 1,207 Bytes
5e2bd8d
 
 
 
 
 
 
 
 
51cb78d
 
5e2bd8d
 
51cb78d
a03684f
 
 
 
51cb78d
 
 
 
 
 
 
 
 
a03684f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
---
library_name: transformers
base_model: p-e-w/gemma-4-E2B-it-heretic-ara
tags:
- onnx
- webgpu
- transformers
- transformers.js
- quantized
- text
- image
- gemma4
---
<!-- AUTOGENERATED HERETIC-TO-ONNX MODEL CARD -->
# 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](https://huggingface.co/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.