How to configure VRAM/main memory split on the GMKtec EVO-X2 (Strix Halo) for local LLM inference. A 29.6GB model ran fine with just 8GB of dedicated VRAM.
Building an NSFW-capable local LLM on the GMKtec EVO-X2 (Strix Halo). Getting GPU inference at ~11 tokens/s with LM Studio and MS3.2-24B-Magnum-Diamond.
Upscaling images loaded via the Load Image node was producing garbled output. Fixed it by addressing the non-contiguous tensor issue — a one-line patch to comfy/utils.py. Added a 2026-04-29 follow-up after a ComfyUI update wiped the patch and the bug came back, with the upstream PyTorch issue and a recurrence-detection snippet.
Went 0-for-13 trying to train an Illustrious-XL LoRA on a Mac Studio M1 Max 64GB. With help from multiple AI agents, pinpointed the root causes and finally succeeded on a RunPod RTX 4090. The full record: three fatal parameters and the sd-scripts trap.
A thinking block corruption error crashed my Claude Code Web session and wiped uncommitted files. An investigation into the cause, workarounds, and whether the files can be recovered.
Running Phr00t's Qwen-Rapid-AIO-NSFW v18.1 (28GB) on RunPod: RTX 4090 24GB froze on load and --lowvram didn't help, the FP8 split needs separate VAE/text encoder, RTX 5090 32GB ran end-to-end.
When building a WebRTC voice call, you can't pass a remote MediaStream to the SpeechRecognition API. Here are three workable approaches — remote-side recognition, server-side processing, and AudioContext — plus iOS-specific implementation strategies.
Generalized the scripts from the practice and optimization articles into a reusable framework and published it on GitHub. A walkthrough of how to use it and the design philosophy.
The Web Speech API + Gemini + VOICEVOX setup is complete — an AI character you can actually have a voice conversation with. Key implementation notes and impressions.
