z-lab/Qwen3.5-4B-DFlash
Different architecture vs Qwen/Qwen3.5-4B. Different architecture: only 34 of 738/69 tensors match by name and shape.
9 derivatives checked
- AtomicChat/Qwen3.5-4B-DFlash-GGUF: Quantization
- Anbeeld/Qwen3.5-4B-DFlash-GGUF: Same weights, converted
- magnitudedev/Qwen3.5-4B-DFlash-GGUF: Quantization
- yugeshkarunamurthy/Qwen3.5-4b-Dflash-6bit-MLX: Quantization
- blake-lucas/Qwen3.5-4B-DFlash-W4A16-MSE-G32-BF16-QKV: Quantization
- blake-lucas/Qwen3.5-4B-DFlash-W4A16-RTN: Quantization
- shujunyi/Qwen3.5-4B-DFlash-Q4_0-GGUF: Quantization
- EntityDeletr/Qwen3.5-4B-DFlash-GGUF: Identical copy
- naveenrajk/Qwen3.5-4B-DFlash-W8A16: Quantization
2 other repos hold byte-for-byte identical weight files.