Editor’s picks
All stories
Guides
Which Qwen3.8 quant belongs on your GPU?
Seven actual GGUF downloads show what GSQ-RCO, IQ3_S, Q3_K_S and ternary packing buy you on a 24 GB card.

AI & Models
Qwen3.8-27B makes the case for a local coding upgrade
Same parameter count, stronger reported coding results. What changes when you move from Qwen3.6 to 3.8 on a 24 GB GPU?

Self-Hosting
Immich fixes two quiet failures in library monitoring
The 3.4.2 patch repairs integrity-report downloads and the Prometheus exporter. Both matter when checking on a photo archive.
How much memory do model weights need?
Compare actual GGUF downloads
| Build | File size | Source |
|---|---|---|
| GSQ-RCO IQ2_XS | npario | |
| GSQ-RCO IQ2_S | npario | |
| GSQ-RCO IQ3_XXS | npario | |
| GSQ-RCO IQ3_S | npario | |
| Q3_K_S | bartowski | |
| Ternary Q2_0 | oshiteku | |
| Ternary PTQ1_0 | oshiteku |
Choose your local AI stack
Compare agent workflows, chat interfaces and inference runtimes. Or compare the model weights that run underneath them.
From the lab
How much of a model is actually in VRAM?Field guides
All guides
Guides
Run your first service with Docker Compose
Serve a page from a folder on your computer. Replace the container and prove the page survives.

Guides
Open a local web app privately from your phone
Connect a loopback-only service to an existing tailnet and verify access over cellular.

Guides
Try a new MCP server without guessing what your client supports
Use the July 2026 protocol revision and a read-only tool to check a client–server pairing.
From the archive
Explore the archive