Vikhrmodels/QVikhr-3-1.7B-Instruction-noreasoning
Full fine-tune vs Qwen/Qwen3-1.7B. Fine-tune: 100% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 1/1, Norms 87/113, MLP 84/84, Attention 112/112
3 derivatives checked
- mradermacher/QVikhr-3-1.7B-Instruction-noreasoning-i1-GGUF: Quantization
- mradermacher/QVikhr-3-1.7B-Instruction-noreasoning-GGUF: Same weights, converted
- DevQuasar/Vikhrmodels.QVikhr-3-1.7B-Instruction-noreasoning-GGUF: Same weights, converted