openjay

openjay

·

openjay

AI & ML interests

None yet

Recent Activity

upvoted a changelog 6 days ago

Filter Models page by Hardware

new activity 3 months ago

mlx-community/gemma-4-e2b-4bit:⚠️ Existing MLX quantized Gemma 4 models (mlx-community, unsloth) produce garbage output due to quantizing PLE (Per-Layer Embedding) layers.

liked a model 9 months ago

Salesforce/Llama-xLAM-2-8b-fc-r

View all activity

Organizations

New activity in mlx-community/gemma-4-e2b-4bit 3 months ago

⚠️ Existing MLX quantized Gemma 4 models (mlx-community, unsloth) produce garbage output due to quantizing PLE (Per-Layer Embedding) layers.

#1 opened 3 months ago by