LuminaBench
Anthropic / Tested model

Claude Fable 5.2

Benchmark results and original creations.

3 tested configurations9 test results · 3 categoriesReleased 9 November 2026
Best recorded results

Benchmark scores

Best eligible score in each benchmark.

Lumina Capabilities Index1,161Max · Expanded · 2/5 · Partial available-category average · Experimental2/5 categories available · 1 fully measured · 4 observed benchmark recordsLumina Voxel Index1,114HighLumina 3D Index1,115HighLumina GameDev Index1,118HighCursorBench1,180Max · ExperimentalARC-AGI-21,132xHigh · ExperimentalGPQA Diamond1,088Max · ExperimentalMystery Game Puzzles1,300Max · ExperimentalFrontierMath Tiers 1–31,119Max · ExperimentalOTIS Mock AIME 2024–20251,073Max · ExperimentalLumina Coding Index1,149Max · 1/2 Estimated · ExperimentalLumina Reasoning Index1,172Max · 3/3 Observed · ExperimentalLumina Mathematics Index1,096Max · 2/2 Observed · Experimental

API pricing

USD / 1M tokens

Verified pricing is unavailable for this model.

Claude Fable 5.2 · High: Voxel Pagoda front view

Voxel

By LuminaDesigned and run by LuminaBench. Original outputs and evidence are retained.

3 configurations

1,114Best score · High · Claude Code
Explore Voxel
Claude Fable 5.2 · High: 3D Seaplane front view

3D

By LuminaDesigned and run by LuminaBench. Original outputs and evidence are retained.

3 configurations

1,115Best score · High · Claude Code
Explore 3D
Claude Fable 5.2 high · six-character selection

GameDev

By LuminaDesigned and run by LuminaBench. Original outputs and evidence are retained.

3 configurations

1,118Best score · High · Claude Code
Explore GameDev
CursorBench

CursorBench

Version 4.0
Lumina VerifiedLumina checked the published sources, versions, model labels, score extraction and conversion. External evaluations were run by their publishers; Lumina has not independently rerun or reproduced them.

5 configurations · Publisher-run · Experimental conversion

1,180Best score · Max
Explore CursorBench
ARC-AGI-2

ARC-AGI-2

Version 2
Lumina VerifiedLumina checked the published sources, versions, model labels, score extraction and conversion. External evaluations were run by their publishers; Lumina has not independently rerun or reproduced them.

5 configurations · Publisher-run · Experimental conversion

1,132Best score · xHigh
Explore ARC-AGI-2
GPQA Diamond

GPQA Diamond

Version Diamond-198
Lumina VerifiedLumina checked the published sources, versions, model labels, score extraction and conversion. External evaluations were run by their publishers; Lumina has not independently rerun or reproduced them.

1 configuration · Publisher-run · Experimental conversion

1,088Best score · Max
Explore GPQA Diamond
Mystery Game Puzzles

Mystery Game Puzzles

Version Puzzles-100
Lumina VerifiedLumina checked the published sources, versions, model labels, score extraction and conversion. External evaluations were run by their publishers; Lumina has not independently rerun or reproduced them.

1 configuration · Publisher-run · Experimental conversion

1,300Best score · Max
Explore Mystery Game Puzzles
FrontierMath Tiers 1–3

FrontierMath Tiers 1–3

Version 2 · Private
Lumina VerifiedLumina checked the published sources, versions, model labels, score extraction and conversion. External evaluations were run by their publishers; Lumina has not independently rerun or reproduced them.

1 configuration · Publisher-run · Experimental conversion

1,119Best score · Max
Explore FrontierMath Tiers 1–3
OTIS Mock AIME 2024–2025

OTIS Mock AIME 2024–2025

Version 2024–2025
Lumina VerifiedLumina checked the published sources, versions, model labels, score extraction and conversion. External evaluations were run by their publishers; Lumina has not independently rerun or reproduced them.

1 configuration · Publisher-run · Experimental conversion

1,073Best score · Max
Explore OTIS Mock AIME 2024–2025

Lumina Coding Index

Version 2
By LuminaIndex designed by LuminaBench using attributed, publisher-run benchmarks. Lumina did not rerun those evaluations.

5 configurations · Derived from published benchmarks · Experimental · 1/2 Estimated

1,149Best score · Max
Explore Lumina Coding Index

Lumina Reasoning Index

Version 3
By LuminaIndex designed by LuminaBench using attributed, publisher-run benchmarks. Lumina did not rerun those evaluations.

5 configurations · Derived from published benchmarks · Experimental · 3/3 Observed

1,172Measured score · Max
Explore Lumina Reasoning Index

Lumina Mathematics Index

Version 3
By LuminaIndex designed by LuminaBench using attributed, publisher-run benchmarks. Lumina did not rerun those evaluations.

1 configuration · Derived from published benchmarks · Experimental · 2/2 Observed

1,096Best score · Max
Explore Lumina Mathematics Index