Skip to content

Search LUMINA BENCH

Navigate the site or open source-checked model and benchmark registry records.

Search models, benchmarks and pages

ESC

Navigate

Source-qualified snapshot · Score 1.7.0↑↓ navigate · Enter open

Latest rankings

Navigation

Lumina Bench

LeaderboardSee the latest model rankings CompareCompare models side by side ModelsBrowse the model directory BenchmarksExplore benchmark evidence

News

AI NewsRead the latest AI stories

Research

How Rankings WorkUnderstand the ranking methodology ChangelogReview ranking and product updates

Latest rankings

Independent rankings. Commercial relationships never affect scoring.

Loading table

Loading LUMINA BENCH

Independent power rankings from hard benchmarks — speed and price stay decision context, not rank inputs.

Product

Leaderboard
Compare
Models
Benchmarks

Research

How rankings work
Changelog

Company

News
Contact

© 2026 Lumina Bench

Home
Models
Claude Opus 5

Anthropic · stable

Claude Opus 5

Anthropic's Opus-class flagship for complex agentic coding and enterprise work, positioned near Fable 5 intelligence at Opus 4.8 pricing.

#1 · 82.9 score · Confidence B

Loading Claude Opus 5 profile