Oxy29/IBM-GPT5.4-Coder-1B
Full fine-tune vs ibm-granite/granite-4.0-1b. Fine-tune: 100% of the weights changed, across 40 of 40 layers.
Tensors changed: Embeddings 1/1, Norms 33/81, Attention 160/160, MLP 80/80
2 derivatives checked
- mradermacher/IBM-GPT5.4-Coder-1B-i1-GGUF: Quantization
- mradermacher/IBM-GPT5.4-Coder-1B-GGUF: Same weights, converted
1 other repos hold byte-for-byte identical weight files.