My DeepSeek-V4.1-Flash-1017 quant is NOW LIVE for TP=2 DGX Sparks. Same 353 questions for all three packs: new / Mia 2.9 / my old q28-v2 MMLU-Pro STEM: 93 / 91 / 90 (of 100) LiveCodeBench: 14 / 11 / 11 (of 17) MBPP+: 50 / 49 / 50 (of 60) HumanEval+: 34 / 34 / 33 (of 40) SciCode: 6 / 6 / 5 (of 10) MATH-500: 36 / 36 / 36 (of 40) MGSM: 40 / 40 / 40 BFCL tool calls: 39 / 39 / 39 (of 40) Multi-step tool chains: 6/6, 6/6, 4/6 These are small samples, so I read it as "at least as good, in a smaller pack". It's not a blowout. The gaps I care about are LiveCodeBench and the tool chains, which my old quant kept dropping. Calibrated it for coding, science and agent work. Weights/Recipe are in the comments. Recipe has been updated with my current config but is mostly unchanged. @MiaAI_lab I still don't know how you get them so good. I can't wait to use your Engine Aware Quant (EAQ) system when it's ready. 2.8 BPW, EXL3, 203.8 GB, running on two DGX Sparks with @ashxhart TensorFold. - 1M context - image input - 2.1M-token cache total with NVMe offload. As usual, if you have any issues, please let me know or submit a PR. What will you throw at it first?

Just wanted to give an update on LocalRouter. I am working on a few new tools and some of the feedback. Nothing is quite ready yet, but it's moving quickly Nearly ready - Jev style decision model for computer use. - Remote companion and choose-your-own harness. - Smaller quant for Dramma thanks to @plotarmordev Also porting my companion mobile app over too, but I want to ask a question. I don't want to deal with App Store submissions right now (Maybe later). Would you prefer an Expo app that you would need an Expo account for, or a raw swift/kotlin build that you can just download to the device directly? Both would just work over Tailscale. Let me know below!