wimmmm/RedHatAI-gemma-4-31B-it-speculator-eagle3-GGUF
Full fine-tune vs RedHatAI/gemma-4-31B-it-speculator.eagle3. wimmmm/RedHatAI-gemma-4-31B-it-speculator-eagle3-GGUF is a GGUF (Q8_0) of different weights. Fine-tune: 100% of the weights changed, across 1 of 1 layers.
Tensors changed: other 2/2, Embeddings 1/1, Norms 0/3, MLP 3/3, Attention 4/4, lm_head 1/1