ibm-granite/granite-3.1-8b-instruct
Full fine-tune vs ibm-granite/granite-3.1-8b-base. Fine-tune: 100% of the weights changed, across 40 of 40 layers.
Tensors changed: Norms 81/81, MLP 120/120, Attention 160/160
18 derivatives checked
- ibm-granite/granite-3.2-8b-instruct: Full fine-tune
- bartowski/granite-3.1-8b-instruct-GGUF: Same weights, converted
- lmstudio-community/granite-3.1-8b-instruct-GGUF: Quantization
- RedHatAI/granite-3.1-8b-instruct-quantized.w4a16: Quantization
- Mungert/granite-3.2-8b-instruct-GGUF: Full fine-tune
- Mungert/granite-3.2-8b-instruct-preview-GGUF: Full fine-tune
- EasierAI/Granite-3.1-8B: Quantization
- mradermacher/granite-3.1-8b-instruct-GGUF: Same weights, converted
- RedHatAI/granite-3.1-8b-instruct-FP8-dynamic: Quantization
- mradermacher/granite-3.1-8b-instruct-i1-GGUF: Quantization
- RedHatAI/granite-3.1-8b-instruct-quantized.w8a8: Quantization
- Alcoft/granite-3.1-8b-instruct-GGUF: Same weights, converted
- ibm-granite/granite-3.2-8b-instruct-preview: Full fine-tune
- tensorblock/granite-3.1-8b-instruct-GGUF: Quantization
- DevQuasar-5/ibm-granite.granite-3.1-8b-instruct-GGUF: Quantization
- ruslanmv/granite-3.1-8b-Reasoning: LoRA-style
- allura-org/Gr3.1-8B-Desert-Rose: LoRA-style
- mindw96/Granite-3.1-8B-instruct-Korean: Full fine-tune
5 other repos hold byte-for-byte identical weight files.