expert-pruning-epi
MoE expert removal measured by EPI -- Qwen3-30B-A3B on Raspberry Pi 5
Placeholder -- full content coming soon. See the GitHub repo for current work.
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Franzabner/expert-pruning-epi", dtype="auto")