Provider 5 · mid tier · low effort · October 2026 (demo)
Demo data · not real resultsModel K · low effort
Navigator Index83.695% CI 79.6–87.4
Rank7–9Ranks 7–9: we can't statistically separate it from the models in that range. Of 14.▼1+3.8
Cost / successful task$0.00443rd cheapest of 14 priced models
Median latency7.0 sPass rate 86% · every time 73%
01Categories
Where it is strong, and where it is not
Category · weightModel K
Coding 15%75.059.2–90.8 · 5 tasks
Data analysis 15%86.778.3–95.1 · 5 tasks
Extraction 10%90.078.8–99.3 · 5 tasks
Reasoning 15%80.873.4–88.3 · 5 tasks
Research & agentic 15%For now these are single-call multi-step tasks: one model call returns plan → steps → deliverable, checked step by step. Tool-using, browsing and long agentic tasks are deferred.87.578.2–95.9 · 5 tasks
Summarization 10%76.758.0–95.3 · 5 tasks
Support 10%80.074.4–85.6 · 5 tasks
Writing 10%94.292.3–96.0 · 5 tasks
Run receipts120 demo runs · no receipts
02Cost
What a working answer costs
$0.0044per successful task
$4.42per 1,000 successes
$442per 100,000 successes
86%runs that pass
120runs measured
4 of 13models it dominates
Cost per successful task divides everything measured on candidate calls by the tasks that passed, so failed runs are priced in. Judge cost is excluded. Definition
Dominates
Significantly better on the Index (paired test) and cheaper per successful task:
03Trend
Across published cycles
Navigator Index — higher is better · shaded: 95% interval
Published cycle
04Receipts
Every run, on the record
Demo data · not real results This is demo data: these runs do not exist, so there are no receipts. On a published cycle this lists every run with its prompt, output, checks and measured cost.
Navigator Bench · Index v2.0 · data through 14 Oct 2026Methodology