google/gemma-4-26B-A4B-it
Full fine-tune vs google/gemma-4-26B-A4B. Fine-tune: 98% of the weights changed, across 30 of 30 layers.
Tensors changed: Embeddings 2/4, MLP 150/231, Norms 211/433, other 59/122, Attention 115/223
278 derivatives checked
- cyankiwi/gemma-4-26B-A4B-it-AWQ-4bit: Quantization
- autotrust/GEV-26B-Decide: Identical copy
- RedHatAI/gemma-4-26B-A4B-it-FP8-dynamic: Quantization
- nvidia/Gemma-4-26B-A4B-NVFP4: Quantization
- unsloth/gemma-4-26B-A4B-it-GGUF: Not comparable yet
- HauhauCS/Gemma4-26B-A4B-QAT-Uncensored-HauhauCS-Balanced-MTP: Full fine-tune
- ggml-org/gemma-4-26B-A4B-it-GGUF: Not comparable yet
- RedHatAI/gemma-4-26B-A4B-it-NVFP4: Quantization
- FINAL-Bench/POCKET-26B-GGUF: Quantization
- ReadyArt/Dark-Scarlett-v0.3-26B-A4B-GGUF: Quantization
- lmstudio-community/gemma-4-26B-A4B-it-MLX-4bit: Quantization
- unsloth/gemma-4-26B-A4B-it-NVFP4: Quantization
- lmstudio-community/gemma-4-26B-A4B-it-MLX-8bit: Quantization
- lmstudio-community/gemma-4-26B-A4B-it-MLX-6bit: Quantization
- lmstudio-community/gemma-4-26B-A4B-it-MLX-5bit: Quantization
- HauhauCS/Gemma4-26B-A4B-Uncensored-HauhauCS-Balanced: Quantization
- Jiunsong/supergemma4-26b-uncensored-gguf-v2: Full fine-tune
- google/gemma-4-26B-A4B-it-qat-q4_0-unquantized: Full fine-tune
- bartowski/google_gemma-4-26B-A4B-it-GGUF: Same weights, converted
- bg-digitalservices/Gemma-4-26B-A4B-it-NVFP4: Quantization
- lmstudio-community/gemma-4-26B-A4B-it-GGUF: Quantization
- LargitData/gemma-4-26b-a4b-it-fp8: Quantization
- unsloth/gemma-4-26B-A4B-it: Identical copy
- mudler/gemma-4-26B-A4B-it-APEX-GGUF: Quantization
- Firworks/gemma-4-26B-A4B-it-fp8: Quantization
- autotrust/JEV-Gemma4-26B-A4B: Identical copy
- cklaus/gemma-4-26B-A4B-it-NVFP4: Quantization
- cyankiwi/gemma-4-26B-A4B-it-AWQ-8bit: Quantization
- tiny-random/gemma-4-moe: Different architecture
- mlx-community/gemma-4-26b-a4b-it-4bit: Quantization
53 other repos hold byte-for-byte identical weight files.