LuminaBench

Latest news

Lumina Journal

Showing 3 of 9 stories

LuminaBench / Capabilities

AI capabilities, compared.

One index across our creation tests. Explore the quality of the results and the resources it takes to make them.

Explore the index

Lumina Capabilities Index

Capability across tested configurations.

LuminaBench
Show
Complete suite · all active categories · 20 of 28 configurations
03006009001,200Claude Opus 5.5 · Extra · Claude Code · 10931,093Claude Opus 5.5ExtraClaude Opus 5.5 · High · Claude Code · 10891,089Claude Opus 5.5HighGPT-6 Astra · Max · Codex · 10851,085GPT-6 AstraMaxClaude Sonnet 5.5 · Extra · Claude Code · 10841,084Claude Sonnet 5.5ExtraGPT-6 Astra · xHigh · Codex · 10791,079GPT-6 AstraxHighClaude Sonnet 5.5 · High · Claude Code · 10771,077Claude Sonnet 5.5HighClaude Opus 5.5 · Medium · Claude Code · 10681,068Claude Opus 5.5MediumClaude Sonnet 5.5 · Medium · Claude Code · 10681,068Claude Sonnet 5.5MediumGPT-6 Astra · High · Codex · 10671,067GPT-6 AstraHighGPT-6.1 Sol · Max · Codex · 10671,067GPT-6.1 SolMaxGPT-6.1 Sol · xHigh · Codex · 10641,064GPT-6.1 SolxHighGPT-6.1 Sol · High · Codex · 10501,050GPT-6.1 SolHighClaude Sonnet 5.5 · Low · Claude Code · 10461,046Claude Sonnet 5.5LowGPT-6 Astra · Medium · Codex · 10421,042GPT-6 AstraMediumGPT-6.1 Sol · Medium · Codex · 10321,032GPT-6.1 SolMediumClaude Opus 5.5 · Low · Claude Code · 10291,029Claude Opus 5.5LowGPT-6 Astra · Light · Codex · 10121,012GPT-6 AstraLightGrok 4.7 · xHigh · Grok Build · 10091,009Grok 4.7xHighClaude Haiku 5.5 · High · Claude Code · 10071,007Claude Haiku 5.5HighGPT-6.1 Sol · Light · Codex · 10061,006GPT-6.1 SolLight

Capabilities Index vs Cost

Capability and total suite cost

LuminaBench
Complete suite · all active categories
$0.03$0.1$0.3$1$3$108009001,0001,100Cost · USD (log scale)Capabilities Index

Capabilities Index vs Generation Time

Capability and total suite generation time

LuminaBench
Complete suite · all active categories
310301003008009001,0001,100Generation time · min (log scale)Capabilities Index

Capabilities Index vs Tokens

Capability and total suite tokens

LuminaBench
Complete suite · all active categories
30K100K300K1M3M10M8009001,0001,100Total tokens (log scale)Capabilities Index

Capabilities Index vs Tool Calls

Capability and total suite tool calls

LuminaBench
Complete suite · all active categories
10301003008009001,0001,100Tool calls (log scale)Capabilities Index

Scores average the category indexes, rounded to a whole number. Resource totals add the completed tests.