🚀 Qwen-MTP Collection ⚡ MTP (Multi Token Prediction) speculative decoding enables models like Qwen3.6 to have ~1.4-2.2x faster generation with no change in accuracy. • 7 items • Updated about 23 hours ago • 25
Jackrong/Negentropy-claude-opus-4.7-9B-GGUF Image-Text-to-Text • 9B • Updated May 8 • 15.2k • 64
KyleHessling1/Qwopus-GLM-18B-Healed-MLX-4bit Text Generation • 16B • Updated Apr 20 • 400 • 14