Observed arrival · 2026-11-15
Wallras: a self-hosted monitor for AI drift
Wallras describes a production monitor that sits between an app and its AI provider, tracking changes in model behavior from the customer’s own servers.
○Open
⊠Login
$Paid
†Ads
✦Pretty
●Pro
◎Niche
◉Human
⚑Risk
ƒJS
- For
- Teams running AI features in production
- Worth noticing
- The example setup changes an OpenAI client's base_url; the page says streaming, tool calls, and rate limits continue unchanged.
Field notes
The described deployment is a Docker container acting as a transparent proxy, with an example that redirects an OpenAI client's base URL to an internal Wallras address. Its listed checks range from response length and refusal rates to confidence and model swaps. The homepage says model comparisons replay past prompts and account for ordinary run-to-run variation; these are product descriptions, not independently verified results.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
○OpenPublic substance visible
⊠LoginAccess appeared gated
$PaidCommerce or pricing visible
✦PrettyNotable craft visible
◎NicheUnusually specific use
One card from the complete issue