AI model leaderboard
The top AI model for text is Claude Fable 5 (59.9 Intelligence Index), the top image model is GPT Image 2 API (1338 Elo), the top video model is Gemini Omni Flash (1326 Elo), and the top voice model is Gemini 3.1 Flash Tts. One leaderboard across every modality — language models by independent benchmarks, media models by human-preference arena Elo.
Language models (LLMs)
Full LLM leaderboard →
Image generation
Full ranking →Video generation
Full ranking →- 1Gemini Omni Flash1326 Elo
- 2Happy Horse 1.1 Text to Video1273 Elo
- 3Seedance 2.0 Fast Text to Video1272 Elo
- 4Seedance 2.0 Mini Text to Video1272 Elo
- 5Seedance 2.0 Text to Video API1272 Elo
- 6Grok Imagine Video1227 Elo
- 7Wan Text to Video1219 Elo
- 8Veo3.1 Lite Text to Video1209 Elo
- 9Veo 3.11206 Elo
- 10Veo 3.1 Fast1204 Elo
Text to speech
Full ranking →Frequently asked
What is the best AI model overall?
For language tasks, Claude Fable 5 leads with a 59.9 Intelligence Index. There is no single "best AI" across modalities — image, video and voice each have their own leader, listed on this page.
How are the different model types ranked?
Language models are ranked by the Artificial Analysis Intelligence Index (independent benchmarks). Image, video and voice models are ranked by human-preference Elo from head-to-head arena votes — the standard for generative quality, where automated benchmarks fall short.
Where do LLM-only rankings live?
The dedicated LLM leaderboard ranks all 300+ language models with coding, agentic, speed and price columns; the /best rankings break it down by use case (coding, design, writing, agents and more).