RefractAI

Fix Gemma inference generation

2

#### Context Running generate.py for Gemma 2B gives the following error: _File "torchtune/modules/kv_cache.py", line 47, in update assert input_pos.shape[0] == k_val.shape[2] AttributeError: 'NoneType' object has no attribute 'shape'_ #### Changelog...

Request: Flux (Diffusion transformer)

3

I anticipate there will be a lot of demand to train (and infer) the new open SOTA image model "Flux". It's the top model on HF right now. It's a...

Request to support the Flux model (T2I diffusion transformer)

1

### 🚀 The feature, motivation and pitch This request is to adapt this to improve the training speed of Flux, a diffusion transformer. It's the top model on HuggingFace trending...