Jev routes Notra chat at about 300 ms p50

Author: Notra

Classification

Notra

GitHub

Classify this use case

  • Classification1.00
  • Shipped0.95
  • Measured0.98
  • Usefulness0.69
  • Repo1.00
  • Ready0.95
  • Specific case0.57
  • Copy quality0.56
  • Creativity0.77
  • Distinct intent0.77
Latency
~300 ms p50 / ~550 ms p95
Result
Boolean answers thresholded at 0.5

Ask Jev a complexity Choice plus tools and reasoning-heavy Nouls on each Notra chat message, then pick a model at a 0.5 threshold.

Questions

Primitive

Choice

Ask a multiple choice question

Name
complexity
Instructions
How complex is the task in userMessage?
Criteria
  • simple

    Single-step task, quick edit, straightforward change, greeting, or simple question

  • complex

    Multi-step task, content creation from scratch, research, analysis, or a task requiring reasoning

Primitive

Noul

Evaluate how true something is

Name
requiresTools
Instructions
Does fulfilling userMessage require using tools (editing the current draft, fetching data from connected sources, accessing skills, publishing or scheduling)?
Criteria
  • true

    Any editing, data fetching from connected sources, skills, or explicit requests to run tools

  • false

    Answering questions, explaining concepts, or conversation

Primitive

Noul

Evaluate how true something is

Name
reasoningHeavy
Instructions
Is the task in userMessage reasoning-heavy?
Criteria
  • true

    Deep analysis, long-form writing, multi-source research, architectural decisions, nuanced critique, comparative synthesis

  • false

    Everyday chat, quick edits, factual lookups, straightforward content generation