DeepSeek V4.1 Flash 1017 on 4x CMP 170HX: 180 tok/s structured with EXL3 experts-only quant
Local deployment test quantizes only routed experts to EXL3 2.9 and keeps official FP8 elsewhere, hitting 180 tok/s structured, 137 code, 97 prose, a 1.53M KV pool and 9/9 needle retrieval at 30K-241K.
DeepSeek V4.1 Flash 1017 4× CMP 170HX ⚡️🔥🎙️ Structured 180 tok/s Code 137 tok/s Prose 97 tok/s Aggregate tok/s, 2 / 4 / 8: Structured 248 / 419 / 621 Code 128 / 173 / 376 Prose 134 / 188 / 283 KV pool 1.53M Routed experts only are EXL3 2.9. Attention, shared experts, head, draft, and Engram stay the official FP8 1017 weights. LibertAI's full 2.9 pack also quantizes attention: +1.8% code perplexity, +0.4% prose vs unquantized. This build quantizes less, so it should sit closer. Needle 9/9 at 30K, 121K, and 241K https://github.com/CurtisAccelerate/DeepSeekv41-Flash-1017-EXL3-CMP170HX