<rss version="2.0"><channel><title>CroZone</title><link>https://stacksignal.chet2191.chatgpt.site</link><description>CroZone editorial feed</description><item><title>Local or cloud? Follow the work.</title><description>Codex or Hermes decides how the job runs. Qwen or Luna decides where the thinking happens. Here’s how to pair them for a code fix, a private document or a daily digest.</description><link>https://stacksignal.chet2191.chatgpt.site/local-vs-cloud.html</link><guid>https://stacksignal.chet2191.chatgpt.site/local-vs-cloud.html</guid></item><item><title>Run your first service with Docker Compose</title><description>Serve a page from a folder on your computer. Replace the container and prove the page survives.</description><link>https://stacksignal.chet2191.chatgpt.site/first-compose-app.html</link><guid>https://stacksignal.chet2191.chatgpt.site/first-compose-app.html</guid></item><item><title>Set up a NAS share you can actually restore</title><description>Build a small Synology share, check who can write to it, and recover yesterday’s file to a second folder.</description><link>https://stacksignal.chet2191.chatgpt.site/nas-first-share.html</link><guid>https://stacksignal.chet2191.chatgpt.site/nas-first-share.html</guid></item><item><title>Open a local web app privately from your phone</title><description>Connect a loopback-only service to an existing tailnet and verify access over cellular.</description><link>https://stacksignal.chet2191.chatgpt.site/tailscale-serve.html</link><guid>https://stacksignal.chet2191.chatgpt.site/tailscale-serve.html</guid></item><item><title>Bring up an Ubuntu VPS and prove you can recover it</title><description>A first-hour walkthrough for a disposable server, from SSH login to a working page and a recovery check.</description><link>https://stacksignal.chet2191.chatgpt.site/first-ubuntu-vps.html</link><guid>https://stacksignal.chet2191.chatgpt.site/first-ubuntu-vps.html</guid></item><item><title>Try a new MCP server without guessing what your client supports</title><description>Use the July 2026 protocol revision and a read-only tool to check a client–server pairing.</description><link>https://stacksignal.chet2191.chatgpt.site/mcp-compatibility.html</link><guid>https://stacksignal.chet2191.chatgpt.site/mcp-compatibility.html</guid></item><item><title>Build a regression test for an AI release classifier</title><description>Score a prompt on labeled examples before changing models or quantization. The test runner uses only Python.</description><link>https://stacksignal.chet2191.chatgpt.site/eval-release-classifier.html</link><guid>https://stacksignal.chet2191.chatgpt.site/eval-release-classifier.html</guid></item><item><title>Make an AI summary earn each citation</title><description>Build a small source-grounded writing workflow that can abstain when the source is silent.</description><link>https://stacksignal.chet2191.chatgpt.site/source-grounded-ai.html</link><guid>https://stacksignal.chet2191.chatgpt.site/source-grounded-ai.html</guid></item><item><title>Which Qwen3.8 quant belongs on your GPU?</title><description>Seven actual GGUF downloads show what GSQ-RCO, IQ3_S, Q3_K_S and ternary packing buy you on a 24 GB card.</description><link>https://stacksignal.chet2191.chatgpt.site/qwen38-quant-guide.html</link><guid>https://stacksignal.chet2191.chatgpt.site/qwen38-quant-guide.html</guid></item><item><title>Hermes or Open WebUI? Choose by the job.</title><description>One can deliver a report while you sleep. The other gives your household a shared place to talk to models. Here’s where each earns its keep.</description><link>https://stacksignal.chet2191.chatgpt.site/hermes-vs-open-webui.html</link><guid>https://stacksignal.chet2191.chatgpt.site/hermes-vs-open-webui.html</guid></item><item><title>Qwen3.8-27B makes the case for a local coding upgrade</title><description>Same parameter count, stronger reported coding results. What changes when you move from Qwen3.6 to 3.8 on a 24 GB GPU?</description><link>https://stacksignal.chet2191.chatgpt.site/qwen38-local.html</link><guid>https://stacksignal.chet2191.chatgpt.site/qwen38-local.html</guid></item><item><title>Hermes ships a patch across its stable channels</title><description>Version 0.23.3 brings a simpler CLI setup and changes to how long tool-running sessions manage their context.</description><link>https://stacksignal.chet2191.chatgpt.site/hermes-0233.html</link><guid>https://stacksignal.chet2191.chatgpt.site/hermes-0233.html</guid></item><item><title>Immich fixes two quiet failures in library monitoring</title><description>The 3.4.2 patch repairs integrity-report downloads and the Prometheus exporter. Both matter when checking on a photo archive.</description><link>https://stacksignal.chet2191.chatgpt.site/immich-342.html</link><guid>https://stacksignal.chet2191.chatgpt.site/immich-342.html</guid></item><item><title>Archive: Three Qwen3 sizes, one useful local experiment</title><description>The original 8B, 14B and 32B dense releases make a controlled comparison easier. Start with one task you can grade.</description><link>https://stacksignal.chet2191.chatgpt.site/qwen3-dense-models.html</link><guid>https://stacksignal.chet2191.chatgpt.site/qwen3-dense-models.html</guid></item><item><title>Map a Docker app’s data before an upgrade</title><description>A short, read-only inventory shows which files survive a container replacement and which still need a backup plan.</description><link>https://stacksignal.chet2191.chatgpt.site/docker-volume-inventory.html</link><guid>https://stacksignal.chet2191.chatgpt.site/docker-volume-inventory.html</guid></item><item><title>A photo backup needs its catalog, too</title><description>Original files and the Immich database answer different recovery questions. Test both with a disposable library.</description><link>https://stacksignal.chet2191.chatgpt.site/immich-restore-drill.html</link><guid>https://stacksignal.chet2191.chatgpt.site/immich-restore-drill.html</guid></item><item><title>How much of a model is actually in VRAM?</title><description>Ollama exposes the loaded model, its quantization and its reported VRAM size in one local request.</description><link>https://stacksignal.chet2191.chatgpt.site/ollama-memory-check.html</link><guid>https://stacksignal.chet2191.chatgpt.site/ollama-memory-check.html</guid></item><item><title>Read the number behind the model name</title><description>Qwen3-32B has 32.8 billion parameters. That difference matters when estimating a tight memory budget.</description><link>https://stacksignal.chet2191.chatgpt.site/qwen3-memory-math.html</link><guid>https://stacksignal.chet2191.chatgpt.site/qwen3-memory-math.html</guid></item><item><title>What Immich meant by a stable release</title><description>The project’s first stable tag set an expectation around compatibility. Its release announcement is worth reading before planning upgrades.</description><link>https://stacksignal.chet2191.chatgpt.site/immich-stable.html</link><guid>https://stacksignal.chet2191.chatgpt.site/immich-stable.html</guid></item></channel></rss>