Tag: ai
-
An alert that could never fire
A critical alert exists to report that authentication on the LLM gateway has been turned off. It sat quietly in a pending state for a while, which read as a scraping problem — plausible, since Prometheus had just been restarted…
-
Why LiteLLM
Every client that wants a local model — the chat interface, the coding assistant, the document classifier, the voice pipeline — talks to one address instead of reaching into the host running the models directly. LiteLLM is what sits at…
-
The gate was recorded. I just didn’t read it.
A model that rendered an animated 3D solar system scored 0.026, close enough to a blank page that the agent was about to write it off. The image told a different story: a recognizable sun, a small red planet, and…
-
What it takes to trust a local model’s score
The lab runs several open-weight language models locally and wants to know which one is actually worth using for a given job. That means a benchmark, and a benchmark means a harness — and the harness has been wrong more…
-
Status: workflow automation live, the gateway console behind SSO
Workflow automation joined the productivity tier, and the local language model gateway’s admin console is now reachable through single sign-on instead of a shared key. A photo library was also started, and stopped almost immediately by something smaller than a…