THE SCOREBOARD FOR WHAT'S NEXTAI + QUANTUM, SCORED. COMPARED. EXPLAINED.
AIQUANTUMSCORE.

VALUE FRONTIER

Performance
per dollar.

Only models with a verified category score and current USD API price appear. Each category stands alone; there is no universal winner.

agents evidence vs standard API price

GPT-6 LunaGPT-5.6 LunaDeepSeek V4.1 FlashGemini 3.5 Flash-LiteGemini 3.8 FlashGemini 3.7 FlashGemini 3.6 FlashDeepSeek V4 ProGrok 4.6Grok 4.5Gemini 3.5 FlashGPT-6 SolClaude Sonnet 5.5GPT-5.6 TerraGemini 3.1 Pro PreviewGPT-5.6 SolClaude Opus 5.5GPT-6 AstraClaude Fable 5.1Blended input/output USD per 1M tokens →

Green points are efficient within this eligible cohort: no other plotted model has both a lower price and an equal or higher score. Prices use the stated 75/25 input/output mix.

ModelCategory scoreBlended priceConfidenceFrontier
GPT-6 Luna20$0.20Limited evidence · 2 independent domainsEfficient in this cohort
GPT-5.6 Luna20$0.45Limited evidence · 2 independent domains—
DeepSeek V4.1 Flash100$0.52Limited evidence · 2 independent domainsEfficient in this cohort
Gemini 3.5 Flash-Lite0$0.85Limited evidence · 1 independent domains—
Gemini 3.8 Flash60$1.50Moderate confidence · 3 independent domains—
Gemini 3.7 Flash20$1.50Limited evidence · 2 independent domains—
Gemini 3.6 Flash20$1.50Limited evidence · 1 independent domains—
DeepSeek V4 Pro40$1.98Limited evidence · 1 independent domains—
Grok 4.660$3.00Moderate confidence · 2 independent domains—
Grok 4.532$3.00Moderate confidence · 2 independent domains—
Gemini 3.5 Flash20$3.38Limited evidence · 1 independent domains—
GPT-6 Sol20$4.00Moderate confidence · 3 independent domains—
Claude Sonnet 5.580$4.00Moderate confidence · 2 independent domains—
GPT-5.6 Terra40$4.50Limited evidence · 2 independent domains—
Gemini 3.1 Pro Preview0$4.50Limited evidence · 1 independent domains—
GPT-5.6 Sol20$8.00Moderate confidence · 3 independent domains—
Claude Opus 5.5100$8.00Moderate confidence · 3 independent domains—
GPT-6 Astra20$20.00Moderate confidence · 4 independent domains—
Claude Fable 5.140$20.00Limited evidence · 2 independent domains—

Median standard API price

2026-09-292026-09-29
  • 2026-09-29: 3 USD / 1M · Median of 24 priced models

One observation; no trend can be inferred.

Qualified performance per dollar

No dated qualified overall scores yet; no trend is shown.

ARC-AGI-3 Standard benchmark frontier

2026-09-292026-09-29
  • 2026-09-29: 62.713 % · gpt-6-astra · ARC-AGI-3 Semi-Private Standard harness best effort 3 Semi-Private · source ↗

One observation; no trend can be inferred.