Sakalti/ultiima-14B

Full fine-tune vs Qwen/Qwen2.5-14B. Fine-tune: 100% of the weights changed, across 48 of 48 layers.

Tensors changed: Embeddings 1/1, Norms 97/97, MLP 144/144, Attention 336/336, lm_head 1/1

3 derivatives checked

• mradermacher/ultiima-14B-i1-GGUF: Quantization
• mradermacher/ultiima-14B-GGUF: Quantization
• Sakalti/ultiima-14b-pass1: Partial fine-tune (fewer layers)

On the Hub · Downloads on Model Pulse