inclusionAI/ZwZ-4B
Full fine-tune vs Qwen/Qwen3-VL-4B-Instruct. Fine-tune: 100% of the weights changed, across 36 of 36 layers.
Tensors changed: Embeddings 4/4, Norms 164/249, MLP 207/220, Attention 240/240
4 derivatives checked
- mradermacher/ZwZ-4B-i1-GGUF: Quantization (fewer layers)
- mradermacher/ZwZ-4B-GGUF: Same weights, converted (fewer layers)
- inclusionAI/ZwZ-4B-GGUF: Same weights, converted (fewer layers)
- noctrex/ZwZ-4B-GGUF: Same weights, converted (fewer layers)