Latest briefing August 28, 2026 morning edition 13 min read
While the tools that run free on hardware you already own have been sitting finished and largely unnoticed for weeks.
Recipes pulled straight out of the briefings, each one with the place it falls over written down next to the reason to try it.
Start at the FastMetal collection and pick the tier your machine can hold.
Where it breaks
Apple Silicon and Python 3.11+ only, so no Windows and no Intel Macs. The 14B is locked to 448 by 832 and 77 frames with no audio track. Its weights are about 15GB, so a 16GB Mac will not hold it no…
The files live at RuneXX/LTX-2.3-Workflows, which despite the name now covers 2.5.
Where it breaks
The author has not shipped dedicated 2.5 workflows yet and tells you to reuse the 2.3 ones, so expect to fix a node or two. Both custom node packs must be on their latest versions for LTX-2 support,…
The conversion recipe is public at john-rocky/hf-to-litertlm, and the finished graphs are at litert-community.
Where it breaks
Slow. Bonsai takes about 64 seconds on an iPhone 17 Pro, klein five minutes on a Pixel 8a. Klein is fixed at 256 by 256 with one reference image, and changing either means re-exporting. Two settings…
The open tools that keep earning their place in the briefing. One appearance is noise; repeat appearances are the signal.
| Repo | What it is | Days | Last |
|---|---|---|---|
| comfyanonymous/ComfyUI | The node graph almost every local model here targets; if you run creative AI on your own… | 2 | Aug 28 |
| Lightricks/LTX-2 | Video and audio generated together in a single pass, and the base for the anonymous… | 2 | Aug 28 |
| hao-ai-lab/FastVideo | Makes video models run on hardware you own, and the source of the three Apple Silicon… | 2 | Aug 28 |
| Lightricks/ComfyUI-LTXVideo | The official nodes and example workflows for running LTX inside ComfyUI (4.1k stars) | 1 | Aug 28 |
| city96/ComfyUI-GGUF | Loads compressed model files in ComfyUI, which is how most people fit large video models… | 1 | Aug 28 |
| google-ai-edge/LiteRT | The runtime behind the three on-device image models (3.3k stars) | 1 | Aug 28 |