Menlo/Jan-nano-128k
Full fine-tune vs Menlo/Jan-nano. Fine-tune: 90% of the weights changed, across 36 of 36 layers.
Tensors changed: Embeddings 0/1, Norms 105/145, MLP 108/108, Attention 144/144
9 derivatives checked
- unsloth/Jan-nano-128k-GGUF: Same weights, converted
- Menlo/Jan-nano-128k-gguf: Quantization
- bartowski/Menlo_Jan-nano-128k-GGUF: Same weights, converted
- gabriellarson/Jan-nano-128k-GGUF: Same weights, converted
- mradermacher/Jan-nano-128k-i1-GGUF: Quantization
- twhitworth/Jan-nano-128k-w8a8-dyn: Quantization
- mradermacher/Jan-nano-128k-GGUF: Same weights, converted
- DavidAU/Qwen3-Jan-Nano-128k-6B-Brainstorm20x: Partial fine-tune (fewer layers)
- fireworks1231/agentic-4b-sft: LoRA-style
1 other repos hold byte-for-byte identical weight files.