Gryphe/Gemma-4-26B-A4B-StyleTune-V2
Same weights, repacked vs google/gemma-4-26B-A4B-it. Same weights in different files (re-saved, re-sharded or new metadata), plus 1 new tensors.
Tensors changed: Embeddings 0/4, MLP 0/231, Norms 0/433, other 0/122, Attention 0/223
14 derivatives checked
- mradermacher/Gemma-4-26B-A4B-StyleTune-V2-i1-GGUF: Quantization
- mradermacher/Gemma-4-26B-A4B-StyleTune-V2-GGUF: Quantization
- alexisStacksCode/Gemma-4-26B-A4B-StyleTune-V2-QAT-GGUF: Full fine-tune
- EmanuelOverride/Gemma-4-26B-A4B-StyleTune-V2-APEX-GGUF: Quantization
- Kraekin/Gemma-4-26B-A4B-StyleTune-V2-Q4_K_M-GGUF: Quantization
- Wiself/gemma-4-26B-A4B-Styletune-V2-Voice: Same weights, converted (fewer layers)
- Gryphe/Pantheon-Reasoning-26B-A4B-1.1-V2: Full fine-tune
- debais/Gemma-4-26B-A4B-StyleTune-V2-APEX-GGUF: Same weights, converted
- SubMaroon/Boulesis-v2.1-26B-A4B: LoRA-style
- SubMaroon/Boulesis-26B-A4B: LoRA-style
- SubMaroon-exp/Boulesis-v2-26B-A4B: LoRA-style
- ailexleon/Gemma-4-26B-A4B-StyleTune-V2-mlx-4Bit: Quantization
- Kraekin/Gemma-4-26B-A4B-StyleTune-V2-Q4_K_S-GGUF: Quantization
- SubMaroon/Gemma-4-26B-A4B-StyleTune-QK-Heretic: LoRA-style