#dgxspark

Day 16 of my 100-experiment local LLM challenge. I generated a full Japanese-vocal theme song for my family's cat with ACE-Step 1.5 on a DGX Spark — words, music, and voice, all local. The twist: the audio itself renders in ~10 seconds; the slow part is the AI "thinking" about the song's structure (~8 minutes).

Day 15 of my 100-experiment local LLM challenge. I turned one photo of my family cat into a short video with two local video AIs on a DGX Spark: LTX-2.3 generated a 4s video WITH a meow sound in 30 seconds; Wan 2.2 took 475 seconds for near-photoreal quality but no audio. Plus a first/last-frame trick that let me design a camera zoom.

Day 13 of my 100-experiment local LLM challenge. I drove a single cat image with human facial motion to make a "talking cat" with LivePortrait, instead of putting my own face on camera. It worked — but the hard part wasn't the animation, it was getting the AI to recognize the cat's face. The fix was a no-compile patch for XPose. All local on a DGX Spark.

AMD Ryzen AI Halo vs NVIDIA DGX Spark 2026: Which 128GB AI Dev Kit Actually Pays Off

Day 12 of my 100-experiment local LLM challenge. I tried to build a line-art LoRA of my own hand-drawn animation by ripping frames out of the video, instead of preparing stills by hand. Single characters came out broken (fused heads, baked-in backgrounds), I fixed most of it by rebuilding the dataset — but "two characters together" was impossible from video, and the only real fix is to hand-draw the stills. All local on a DGX Spark.

Day 11 of my 100-experiment local LLM challenge. I tried to restyle my cat into anime, ukiyo-e, oil painting and more — but the AI kept drawing humans. A single reference photo (IPAdapter) drags the result back to being my actual cat, all running locally on a DGX Spark.

Day 14 of my 100-experiment local LLM challenge. I asked a general-purpose audio LLM (Qwen2-Audio) to guess what a cat wants from its meow, using a labeled research dataset of 440 meows. It scored 23.3% on a 3-choice quiz — below random — and a shuffled-options control revealed it wasn't really listening at all. Meanwhile Whisper transcribed the meows as "Thank you for watching!" All local on a DGX Spark.

Day 2 of my 100-experiment local LLM challenge. LoRA fine-tuning on 22 cat photos with Kohya_ss on a DGX Spark. 13 minutes of training, and "ohwx cat as a chef" now actually gives me my cat as a chef.

Day 10 of my 100-experiment local LLM challenge. I built a v0.1 weather-officer bot that texts me a clothing suggestion every morning and learns whether I run cold or warm from one-tap feedback — all running on my DGX Spark, with nothing personal leaving the machine.

Day 9 of my 100-experiment local LLM challenge. I fed 8 years of one LINE conversation (87k messages) to local models on my DGX Spark — a Japanese sentiment BERT for tone and Qwen2.5-72B to guess turning points — without sending anything to the cloud. The relationship's arc surfaced from message counts and tone alone.

Day 8 of my 100-experiment local LLM challenge. Followed up Day 7 by swapping multi-digit addition for a harder task — predicting the first-break position in nested brackets, depth 2-20 — using the same OpenMythos-style mini model and training pipeline. Tests whether the loop-count peak generalizes across tasks, and whether deeper inputs benefit from more inference loops.

Day 7 of my 100-experiment local LLM challenge. Trained a tiny OpenMythos-style mini model (theoretical reconstruction of the rumored Claude Mythos architecture) on multi-digit addition to test whether more recurrent loops at inference time actually make the model smarter. Adding a fresh data point to the Saunshi / Huginn / Bee picture.