ibm-granite/granite-4.2-30b
Full fine-tune vs ibm-granite/granite-4.1-30b-base. Fine-tune: 100% of the weights changed, across 64 of 64 layers.
Tensors changed: Embeddings 1/1, Norms 129/129, MLP 192/192, Attention 256/256
26 derivatives checked
- ibm-granite/granite-4.2-30b-GGUF: Same weights, converted
- bartowski/granite-4.2-30b-GGUF: Same weights, converted
- ibm-granite/granite-4.2-30b-nvfp4: Quantization
- mradermacher/granite-4.2-30b-i1-GGUF: Quantization
- ibm-granite/granite-4.2-30b-fp8: Quantization
- lmstudio-community/granite-4.2-30b-MLX-8bit: Quantization
- lmstudio-community/granite-4.2-30b-MLX-4bit: Quantization
- lmstudio-community/granite-4.2-30b-GGUF: Quantization
- ibm-granite/granite-4.2-30b-mxfp4: Quantization
- ibm-granite/granite-4.2-30b-q4-mlx: Quantization
- ibm-granite/granite-4.2-30b-q8-mlx: Quantization
- PS4Research/IKn8jwlNamPNIS15: LoRA-style
- ibm-granite/granite-4.2-30b-bf16-mlx: Same weights, repacked
- lmstudio-community/granite-4.2-30b-MLX-6bit: Quantization
- PS4Research/xnfjGuaVd0Dp3J42: LoRA-style
- xCloudinfo/Granite-4.2-30B-xVITA-zhTW-Uncensored-GGUF: Full fine-tune
- vcruz305/granite-4.2-30b-GGUF: Quantization
- PS4Research/9DgDaWTmX85XRwSq: LoRA-style
- mradermacher/granite-4.2-30b-GGUF: Quantization
- xCloudinfo/Granite-4.2-30B-xVITA-zhTW-GGUF: Full fine-tune
- PS4Research/005Kq3yTVMadlDXH: LoRA-style
- armand0e/Granite-4.2-30B-Fable-Distill: LoRA-style
- dr-robot/granite-4.2-30b-mlx-8Bit: Quantization
- vcruz305/granite-4.2-30b-NVFP4: Quantization
- DevQuasar/ibm-granite.granite-4.2-30b-GGUF: Quantization
- ChonkE/Granite_42_30b_Abliterated: LoRA-style
1 other repos hold byte-for-byte identical weight files.