z-lab/Qwen3.8-27B-DFlash2
Different architecture vs Qwen/Qwen3.8-27B. Different architecture: only 28 of 1199/81 tensors match by name and shape.
22 derivatives checked
- naklitechie/Qwen3.8-27B-DFlash2-ternary-bonsai2: Full fine-tune
- tcclaviger/Qwen3.8-27B-DFlash2-FP8: Quantization
- lued/Qwen3.8-27B-DFlash2-W8: Quantization
- analogalok/Qwen3.8-27B-DFlash2-Q2_K-GGUF: Quantization
- agentionai/Qwen3.8-27B-DFlash2-ROCmFP4-FAST-GGUF: Same weights, converted
- 0xWhiteMage/Qwen3.8-27B-Kearuga-DFlash2: Quantization
- jfan/Qwen3.8-27B-heretic-dflash2: Full fine-tune
- Schiltmans/Ternary-Bonsai-2-27B-DFlash2-ft5: Partial fine-tune
- NewHorizonGroup/Qwen3.8-27B-DFlash2-oQ4: Quantization
- costanzopadovano/Qwen3.8-27B-DFlash2-NVFP4-GGUF: Same weights, converted
- JonasLoos/Qwen3.8-27B-DFlash2-b32: Full fine-tune
- YourHighnessLA/Qwen3.8-27B-DFlash2-NVFP4: Quantization
- JRamirez-UAB/Qwen3.8-27B-DFlash2-W4A16: Quantization
- vcruz305/Qwen3.8-27B-DFlash2-SAGE-EXL3-5.0bpw: Same weights, repacked
- curvedinf/Qwen3.8-27B-DFlash2-GPTQ-INT8-W8A8-GS128: Quantization
- deepsweet/Qwen3.8-27B-DFlash2-FP16: Same weights, converted
- josch15366/Qwen3.8-27B-DFlash2-FP8: Quantization
- 0xMiami305/Qwen3.8-27B-DFlash2-W4A16-MLP-only: Quantization
- 0xBakeer/TandemLLM-Qwen3.8-27B-DFlash2-b16: Full fine-tune
- curvedinf/Qwen3.8-27B-DFlash2-GPTQ-INT8-W8A8-GS128-PTQR: Quantization
- 0xBakeer/TandemLLM-Qwen3.8-27B-DFlash2-b8: Full fine-tune
- Twu31/Qwen3.8-27B-DFlash2-INT8-LowLatency: Quantization
5 other repos hold byte-for-byte identical weight files.