nvidia/Llama-3.1-Nemotron-Nano-4B-v1.1
Full fine-tune vs nvidia/Llama-3.1-Minitron-4B-Width-Base. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: Embeddings 1/1, Norms 65/65, MLP 96/96, Attention 128/128, lm_head 1/1
14 derivatives checked
- bartowski/nvidia_Llama-3.1-Nemotron-Nano-4B-v1.1-GGUF: Same weights, converted
- EQTYLab/Llama-3.1-Nemotron-Nano-4B-v1.1-banking77: LoRA-style
- mradermacher/Llama-3.1-Nemotron-Nano-4B-v1.1-i1-GGUF: Quantization
- redponike/Llama-3.1-Nemotron-Nano-4B-v1.1-GGUF: Same weights, converted
- unsloth/Llama-3.1-Nemotron-Nano-4B-v1.1-unsloth-bnb-4bit: Quantization
- mlx-community/Llama-3.1-Nemotron-Nano-4B-v1.1-4bit: Quantization
- lmstudio-community/Llama-3.1-Nemotron-Nano-4B-v1.1-GGUF: Quantization
- DevQuasar/nvidia.Llama-3.1-Nemotron-Nano-4B-v1.1-GGUF: Same weights, converted
- mradermacher/Llama-3.1-Nemotron-Nano-4B-v1.1-GGUF: Same weights, converted
- mlx-community/Llama-3.1-Nemotron-Nano-4B-v1.1-bf16: Same weights, repacked
- mlx-community/Llama-3.1-Nemotron-Nano-4B-v1.1-8bit: Quantization
- unsloth/Llama-3.1-Nemotron-Nano-4B-v1.1-bnb-4bit: Quantization
- unsloth/Llama-3.1-Nemotron-Nano-4B-v1.1: Identical copy
- Montecarlo2024/Llama3.1-8B-Nemotron-Python-v2: LoRA-style
3 other repos hold byte-for-byte identical weight files.