Immac/NetaYume-Lumina-Image-2.0-GGUF

Full fine-tune (fewer layers) vs duongve/NetaYume-Lumina-Image-2.0. Immac/NetaYume-Lumina-Image-2.0-GGUF matches part of duongve/NetaYume-Lumina-Image-2.0 (400 of its 934 tensors; the rest of duongve/NetaYume-Lumina-Image-2.0 has no counterpart). On the shared part: Immac/NetaYume-Lumina-Image-2.0-GGUF is a GGUF (F16) of different weights. Fine-tune: 99% of the weights changed, across 26 of 26 layers.

Tensors changed: Embeddings 9/9, Norms 141/181, Attention 59/60, MLP 90/90, other 59/60

On the Hub · Downloads on Model Pulse