[Announcement] On hiatus until August 21

The open models are landing thick and fast right now, and my gallbladder picked exactly this moment to give out. Nothing new here for a couple of weeks.
Hi everyone. This is a genuinely great stretch to be into local models — the full DeepSeek V4 Flash, Qwen 3.8, MiniMax H3, all within days of each other. There is a ridiculous amount to play with. Go enjoy it.
I'd love to be joining in. Instead I've spent about a week in a hospital bed: gallstones, a blocked duct, and the kind of inflammation that gets you admitted rather than sent home with painkillers. I kept telling myself I could still get some benchmarking done between the bad stretches.

I can't even.
(Anyone who can actually run benchmarks through this is built differently. XD)
Look after yourselves out there.
Read next
- 2026-11-10How to run Viggle Remake character swap on an RTX 5090 (int8, ComfyUI)
How to set up Viggle Remake in ComfyUI on a single 32 GB RTX 5090 with int8 weights: character swap, scene change, and an 89 vs 74 blind win over a SAM mask + H3 Character Swap pipeline.
- 2026-11-09[AI Workflow] A Self-Running AI News Shorts Pipeline on Local Models: Harness Changes Cut Director Time From 80.6 to 5–10 Minutes
My pipeline makes Chinese and English news shorts 4 times a day. A new harness took the local DS4.1 director from 80.6 to 5–10 minutes per video.
- 2026-11-09[Benchmark] No GPU? Run MiniMax-H3 on a Colab G4: 66 s per clip, ~49 clips an hour
MiniMax-H3 Ref2VA on a Colab G4: 66.1 s per 10-second two-character dialogue clip vs 82–88 s on my RTX 5090, about 49 clips an hour for ~10 compute units.
- 2026-11-04[DGX Spark] DGX Spark and GX10 prices in Taiwan: up 80% in a year, flat since August
GB10 prices in Taiwan, Nov 2025 to Nov 2026: ASUS GX10, MSI EdgeXpert and DGX Spark up 47-80%. Cheapest ways to buy, the 64GB model, RTX Spark vs GB10.
Don't miss the next one
Subscribe, and you won't.
One-click unsubscribe anytime.