ibm-granite/granite-4.2-8b
Full fine-tune vs ibm-granite/granite-4.1-8b-base. Fine-tune: 100% of the weights changed, across 40 of 40 layers.
Tensors changed: Embeddings 1/1, Norms 79/81, MLP 120/120, Attention 160/160
29 derivatives checked
- ibm-granite/granite-4.2-8b-GGUF: Same weights, converted
- bartowski/granite-4.2-8b-GGUF: Same weights, converted
- lmstudio-community/granite-4.2-8b-MLX-8bit: Quantization
- lmstudio-community/granite-4.2-8b-MLX-4bit: Quantization
- ibm-granite/granite-4.2-8b-fp8: Quantization
- lmstudio-community/granite-4.2-8b-GGUF: Quantization
- ibm-granite/granite-4.2-8b-nvfp4: Quantization
- saidutta69/granite-4.2-8b-heretic: Abliterated
- mradermacher/granite-4.2-8b-i1-GGUF: Quantization
- richardyoung/granite-4.2-8b-heretic: Abliterated
- ibm-granite/granite-4.2-8b-mxfp4: Quantization
- NANI-Nithin/granite-4.2-8b-GGUF: Same weights, converted
- webmp3/Sakura-granite-4.2-8b-GGUF: Quantization
- lmstudio-community/granite-4.2-8b-MLX-6bit: Quantization
- ibm-granite/granite-4.2-8b-q4-mlx: Quantization
- xCloudinfo/Granite-4.2-8B-Uncensored-GGUF: Quantization
- ibm-granite/granite-4.2-8b-q8-mlx: Quantization
- ibm-granite/granite-4.2-8b-bf16-mlx: Same weights, repacked
- Jeethu/granite-4.2-8b-PARO: Quantization
- mradermacher/granite-4.2-8b-GGUF: Same weights, converted
- rolandwhere/granite-4.2-8b-symbolic-cot: LoRA-style
- mohameddalii/coda-llm: LoRA-style
- liodon-ai/granite-4.2-8b-imatrix-GGUF: Quantization
- webAI-Official/granite-4.2-8B: Same weights, converted
- salohcin714/granite-4.2-8b-bf16-mlx: Same weights, repacked
- DevQuasar/ibm-granite.granite-4.2-8b-GGUF: Quantization
- salohcin714/granite-4.2-8b-2bit-gptq-mlx: Quantization
- Fr4nkSt3in/granite-4.2-8b-BF16-GGUF: Same weights, converted
- qcokvd1/granite-4.2-8b-decensored: Abliterated
1 other repos hold byte-for-byte identical weight files.