Observed arrival · 2026-10-16
Eden Router puts model choice on a cost-quality dial
Eden Router shows how it would choose a model and reasoning effort for each task, with an interactive dial comparing cost and benchmark results.
○Open
⊠Login
$Paid
†Ads
✦Pretty
●Pro
◎Niche
◉Human
⚑Risk
ƒJS
- For
- Developers balancing model quality and API cost
- Worth noticing
- On this 44-task set, GPT-6.1 Sol passed every task for less than the router’s Level 3 cost.
Field notes
The routing description starts with rules that read the prompt; an optional small model can rate task difficulty. The displayed test covers 18 models from Anthropic, OpenAI, Google, and Moonshot, with 38 rule-graded tasks and six judged by an AI model. The page also concedes that the sample is small, each task ran once, and its task mix leans toward exact computation and code.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
○OpenPublic substance visible
✦PrettyNotable craft visible
●ProPolished or operationally mature
◎NicheUnusually specific use
One card from the complete issue