All LLMs

Top LLMs by overall performance

Models across tasks
913
Publishers
95
VRAM calculator available
370
Sort models
Data as of

Browse models

Models with overall scores 65

Ranked by the Artificial Analysis Intelligence Index among models in this directory. Ties share a rank. Benchmark methodology ↗

API rates are OpenRouter snapshot prices per million input and output tokens. Modalities reflect the listed endpoint or local configuration.
ModelWeight paramsContext windowModalitiesRelease dateIndexAPI price / 1M
#1Claude Fable 5.2—1M2026-11-0964.6
#2GPT-6.2 Astra—1.05M2026-11-1762.1
#3Gemini 4 Neon—1.05M—61.4
#4Gemini 4 Carbon—1.05M2026-11-1060.0
#5Claude Opus 5.5—1M—59.1
#6GPT-6.2 Sol—1.05M—57.3
#7GPT-6 Astra—1.05M2026-09-0356.9
#8GPT-6.1 Sol—1.05M2026-09-2954.6
#9Claude Sonnet 5.5—1M2026-09-2854.5
#10Gemini 4 Hydrogen—1.05M2026-11-1253.9
#11GPT-6.1 Terra—1.05M2026-10-2651.3
#12Gemini 4 Argon—1.05M2026-10-1350.4
#13Claude Fable 5.1—1M—50.0
#14Kimi K3.1 Pro2.78T1.05M2026-10-2346.5
#15Claude Opus 5—1M—44.2
#15GPT-6 Sol—1.05M2026-09-2244.2
#17Muse Wonder—1.05M2026-10-1243.6
#18Qwen4 Max—1M—42.9
#19Muse Spark 1.3—1.05M2026-09-0236.6
#20Claude Haiku 5.5—1M2026-10-0736.5
Estimate VRAM by model (370)