Gryphe/Gemma-4-31B-StyleTune
Same weights, repacked vs google/gemma-4-31B-it. Same weights in different files (re-saved, re-sharded or new metadata), plus 1 new tensors.
Tensors changed: Embeddings 0/4, Norms 0/523, other 0/62, MLP 0/261, Attention 0/338
10 derivatives checked
- bartowski/Gryphe_Gemma-4-31B-StyleTune-GGUF: Same weights, converted (fewer layers)
- mradermacher/Gemma-4-31B-StyleTune-i1-GGUF: Quantization (fewer layers)
- mradermacher/Gemma-4-31B-StyleTune-GGUF: Quantization (fewer layers)
- Wiself/gemma-4-31B-QAT-Styletune-Voice: Full fine-tune
- densenet/Gemma-4-31B-StyleTune-heretic-ara-GGUF: Abliterated (fewer layers)
- Wiself/gemma-4-31B-Styletune-Voice: Same weights, repacked (fewer layers)
- Gryphe/Pantheon-Reasoning-31B-1.1: LoRA-style
- densenet/Gemma-4-31B-StyleTune-heretic-ara: Abliterated
- Casual-Autopsy/G4-MeroMero-31B-StyleSwap: LoRA-style
- Iloqt/Versi-StyleTune-31B: LoRA-style