ibm-granite/granite-3.1-2b-instruct
Full fine-tune vs ibm-granite/granite-3.1-2b-base. Fine-tune: 100% of the weights changed, across 40 of 40 layers.
Tensors changed: Norms 81/81, MLP 120/120, Attention 160/160
17 derivatives checked
- RedHatAI/granite-3.1-2b-instruct-quantized.w4a16: Quantization
- ibm-granite/granite-3.2-2b-instruct: Full fine-tune
- ibm-granite/granite-vision-3.2-2b: Full fine-tune (fewer layers)
- bartowski/granite-3.1-2b-instruct-GGUF: Same weights, converted
- mradermacher/granite-3.1-2b-instruct-i1-GGUF: Quantization
- ibm-granite/granite-vision-3.1-2b-preview: Full fine-tune (fewer layers)
- RedHatAI/granite-3.2-2b-instruct: Full fine-tune
- Mungert/granite-3.2-2b-instruct-GGUF: Full fine-tune
- mradermacher/granite-3.1-2b-instruct-GGUF: Same weights, converted
- Dzluck/granite-3.1-2b-instruct-GGUF: Same weights, converted
- Alcoft/granite-3.1-2b-instruct-GGUF: Same weights, converted
- lmstudio-community/granite-3.1-2b-instruct-GGUF: Quantization
- DevQuasar-6/ibm-granite.granite-3.1-2b-instruct-GGUF: Quantization
- ArRENCEAI/granite-3.1-2b-instruct-OBLITERATED: Full fine-tune
- Prithwiraj731/Granite-3.1-2b-TwoWheeler: LoRA-style
- Prithwiraj731/Granite-3.1-2b-FourWheeler: LoRA-style
- ruslanmv/granite-3.1-2b-Reasoning: LoRA-style
1 other repos hold byte-for-byte identical weight files.