ukisai/Swift-Qwen3.8-27b
Image-Text-to-Text • 28B • Updated • 22.2k • 609
Swift 27B
Swift 1.0, UkisAI's original reasoning-efficient model: 58% fewer thinking tokens than Qwen3.8-27B. BF16 weights and every quant.
Note Swift 1.0. Full BF16 weights. Start here.
ukisai/Swift-Qwen3.8-27B-GGUF
Image-Text-to-Text • 27B • Updated • 344k • 449Note Swift 1.0 GGUF for llama.cpp, LM Studio and Ollama.
ukisai/Swift-Qwen3.8-27B-NVFP4
Image-Text-to-Text • 18B • Updated • 16k • 44Note Swift 1.0, NVFP4/FP8 (NVIDIA ModelOpt) for Blackwell GPUs.
ukisai/Swift-Qwen3.8-27b-BF16-AMD
Image-Text-to-Text • 28B • Updated • 196 • 2Note Swift 1.0, BF16 companion for AMD / ROCm and AMD Quark workflows.
ukisai/Swift-Qwen3.8-27b-int4-AMD
Image-Text-to-Text • 6B • Updated • 1.14k • 8Note Swift 1.0, AMD Quark AWQ INT4 W4A16 for AMD GPUs.