TheDrummer/Orion-26B-A4B-v1
Different architecture vs google/gemma-4-31B-it. Different architecture: only 445 of 1188/1013 tensors match by name and shape.
6 derivatives checked
- bartowski/TheDrummer_Orion-26B-A4B-v1-GGUF: Same weights, converted
- mradermacher/Orion-26B-A4B-v1-i1-GGUF: Quantization
- mradermacher/Orion-26B-A4B-v1-GGUF: Quantization
- K4r14/Orion-26B-A4B-v1-APEX.i-Quality.Q6_K-GGUF: Quantization
- flukethoughts/Orion-26B-A4B-v1-NVFP4: Quantization
- ailexleon/Orion-26B-A4B-v1-mlx-4Bit: Quantization