DMX-compressed shard 6 of the Wan 2.2 I2V 14B model using BFP compression.
This is a single shard demo -- full model compression coming soon.
1git clone https://github.com/willjriley/dmx
2python dmx/dmx_cli.py decompress wan22_shard6.dmx shard6.safetensors
DMX achieves 60-80% compression on neural network weights. Patent Pending.
https://github.com/willjriley/dmx