Leaderboard
The top 10 on April 30, 2025
Ranked by the Artificial Analysis Intelligence Index, with each model's price on that day (per million tokens, 3 parts input to 1 part output).
- 1o3ClosedOpenAI · released Apr 16, 2025 · 2 weeks oldIntelligence 20.2$18 /Mnow $3.50
- 2Claude 3.7 SonnetClosedAnthropic · released Feb 24, 2025 · 2 months oldIntelligence 17.7$6 /M
- 3o4-miniClosedOpenAI · released Apr 16, 2025 · 2 weeks oldIntelligence 16.7$1.93 /M
- 4o1ClosedOpenAI · released Dec 5, 2024 · 5 months oldIntelligence 15.2$26 /M
- 5Gemini 2.5 Pro Preview (Mar' 25)ClosedGoogle · released Mar 25, 2025 · 5 weeks oldIntelligence 15No public price
- 6Grok 3 mini ReasoningClosedxAI · released Feb 19, 2025 · 2 months oldIntelligence 14.6$0.35 /M
- 7GPT-4.1ClosedOpenAI · released Apr 14, 2025 · 2 weeks oldIntelligence 12.7$3.50 /M
- 8o3-miniClosedOpenAI · released Jan 31, 2025 · 3 months oldIntelligence 12.5$1.93 /M
- 9o1-proClosedOpenAI · released Mar 19, 2025 · 6 weeks oldIntelligence 12.4$263 /M
- 10Grok 3ClosedxAI · released Feb 19, 2025 · 2 months oldIntelligence 12.1$6 /M
18 months later
What has happened since
342 models have come out since. Today's best is Claude Opus 5.5 at 57.6, 37.4 points higher.
Hy3-preview scores at least as well today for $0.10 per million tokens, against $18 for o3 back then.
GPT-5 mini matched it for $0.69 per million tokens on Aug 7, 2025.
Kimi K2 Thinking from Kimi matched it on Nov 6, 2025, free to download and run.
The race
Which lab held the top spot
OpenAI led for 2.5 years of the 4 years since ChatGPT launched. Anthropic has held #1 since Apr 16, 2026, 6 models in a row. We rank by the Artificial Analysis Intelligence Index. Other leaderboards, such as LMArena's human votes, crown different models.
The wider field
158 models were out on Apr 30, 2025
DeepSeek, scoring 11.4. #14 overall, 8.8 points behind.
xAI, scoring 14.6. #6 overall at $0.35, 50× cheaper than #1.
- Nova Premier9.2
- Qwen3 235B A22B9.5
- Qwen3 30B A3B7.6
- Qwen3 32B7.3
- Qwen3 4B7.2
- Qwen3 14B6.7
Browse
The top AI model, month by month
Try today's top models on your own prompt
Scores tell you who leads on average. Send one prompt to up to six models and see which answer you'd actually use, with what each one cost. For the long view of prices, read Six months later, the best AI costs 20× less.
Scores and release dates from Artificial Analysis (Intelligence Index, best published effort of each model), data through October 18, 2026. Prices are list prices on the day, including later cuts we know of. Models without a public price are ranked but not priced.