ddalcu/Qwen3.8-Flash-Next-MLX-Serve-mixed-4-8bit Text Generation • 133B • Updated 16 days ago • 13.6k • 53
view article Article 24/24 Retrieval, Yet the Embeddings Changed: An MLX Q4–Q8 Sweep with CUDA Controls TiGa-RCE • Jul 29 • 2