DeepSeek V4 Flash 284B (13B active)

User reports running DeepSeek V4 Flash 0731 with UD-Q4_K_XL quant on 2x RTX 3090 + 1x RTX 3060 (128GB RAM, ~60GB VRAM) using a custom llama.cpp fork. Generation speed is low to mid twenties t/s, prompt processing around 45 t/s with the moe-cache branch, and near 200 t/s with a separate IQ2_M prompt-processing model. Setup uses 167936 context, q8_0 KV cache, dflash speculative decoding, and MoE expert caching with mlock. User notes prompt processing remains a bottleneck and overall prompt handling only improves for prompts over 30K tokens.