Casual-Autopsy/Giftige-Blume-31B-v1-StyleSwap
Same weights, repacked vs Blazed-Forge/Gemma-4-Giftige-Blume-31B-v1. Same weights in different files (re-saved, re-sharded or new metadata), plus 1 new tensors.
Tensors changed: Embeddings 0/3, Norms 0/523, other 0/62, MLP 0/261, Attention 0/338
4 derivatives checked
- mradermacher/Giftige-Blume-31B-v1-StyleSwap-i1-GGUF: Quantization (fewer layers)
- mradermacher/Giftige-Blume-31B-v1-StyleSwap-GGUF: Quantization (fewer layers)
- Rada2025/Gemma-4-31b-RP-MERGE-NVFP4-BF16-GGUF: Quantization
- Casual-Autopsy/Giftige-Blume-31B-v1-StyleSwap-GGUFs: Same weights, converted (fewer layers)