OpenResearcher/OpenResearcher-30B-A3B-GGUF
Full fine-tune vs nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-Base-BF16. OpenResearcher/OpenResearcher-30B-A3B-GGUF is a GGUF (BF16) of different weights. Fine-tune: 100% of the weights changed, across 37 of 52 layers.
Tensors changed: Embeddings 1/1, Norms 38/53, lm_head 1/1