Sakalti/Mystica1-4B

Same weights, converted vs unsloth/Qwen3-4B. Same weights stored in another dtype (BF16 → F16).

Tensors changed: Embeddings 0/1, Norms 0/145, MLP 0/108, Attention 0/144

2 derivatives checked

• mradermacher/Mystica1-4B-i1-GGUF: Quantization
• mradermacher/Mystica1-4B-GGUF: Same weights, converted

1 other repos hold byte-for-byte identical weight files.

On the Hub · Downloads on Model Pulse