EssentialAI/rnj-1-instruct
Full fine-tune vs EssentialAI/rnj-1. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: Embeddings 1/1, Norms 193/193, MLP 96/96, Attention 128/128
10 derivatives checked
- lmstudio-community/rnj-1-instruct-GGUF: Quantization
- unsloth/rnj-1-instruct-GGUF: Same weights, converted
- bartowski/EssentialAI_rnj-1-instruct-GGUF: Same weights, converted
- DevQuasar/EssentialAI.rnj-1-instruct-GGUF: Same weights, converted
- noctrex/rnj-1-instruct-GGUF: Same weights, converted
- prithivMLmods/rnj-1-instruct-GGUF: Same weights, converted
- lmstudio-community/rnj-1-instruct-MLX-8bit: Quantization
- kalomaze/hex-toolcall-base-v2: Full fine-tune
- kalomaze/findfile-xml-step240: Partial fine-tune
- vanta-research/scout-8b: LoRA-style
1 other repos hold byte-for-byte identical weight files.