incoai/Qwen3.8-27B-DFlash2
Different architecture vs Qwen/Qwen3.8-27B. Different architecture: only 28 of 1199/81 tensors match by name and shape.
12 derivatives checked
- maurienne-ai/Qwen3.8-27B-DFlash2-NVFP4-RTNcal: Quantization
- syvai/Qwen3.8-27B-DFlash2-W4A16: Quantization
- HermiHg/Qwen3.8-27B-DFlash2-Q2_K_S-MIX-GGUF: Quantization
- Anbeeld/Qwen3.8-27B-DFlash2-GGUF: Same weights, converted
- andrew-paul/Qwen3.8-27B-DFlash2-Q3_K_M-GGUF: Quantization
- deresolution/Qwen3.8-27B-DFlash2-mxfp4: Quantization
- r0b0tlab/Qwen3.8-27B-DFlash2-EXL3-4.00bpw: Same weights, converted
- hamichok/Qwen3.8-27B-DFlash2-NVFP4-modelopt: Quantization
- doth4580/Qwen3.8-27B-DFlash2-NVFP4-RTNcal-FIXED: Same weights, repacked (fewer layers)
- BlairQ/qwen3.8_27b_idiotSavant_sm_86_dflash2: Quantization
- Busdriver/Qwen3.8-27B-DFlash2-Marathon-GGUF: Quantization
- gratex/Qwen3.8-27B-DFlash2-W4A16-g128-sym-GPTQ: Quantization
5 other repos hold byte-for-byte identical weight files.