Leaderboard
The top 10 on July 31, 2024
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). The index uses today's harder tests, so models from before late 2024 all score between 5 and 9 and their order is rough.
- 1Claude 3 OpusClosedAnthropic · released Mar 4, 2024 · 5 months oldIntelligence 8.7$30 /M
- 2GPT-4o (May '24)ClosedOpenAI · released May 13, 2024 · 3 months oldIntelligence 7.3$7.50 /M
- 3Llama 3.1 Instruct 405BOpen weightsMeta · released Jul 23, 2024 · 8 days oldIntelligence 7.3No public price
- 4Claude 3.5 Sonnet (June '24)ClosedAnthropic · released Jun 21, 2024 · 6 weeks oldIntelligence 7.2$6 /M
- 5GPT-4 TurboClosedOpenAI · released Nov 6, 2023 · 9 months oldIntelligence 7$15 /M
- 6Llama 3.1 Instruct 8BOpen weightsMeta · released Jul 23, 2024 · 8 days oldIntelligence 6.9$0.04 /M
- 7Mistral Large 2 (Jul '24)ClosedMistral · released Jul 24, 2024 · 7 days oldIntelligence 6.8$3 /M
- 8GPT-4ClosedOpenAI · released Mar 14, 2023 · 17 months oldIntelligence 6.7$38 /M
- 9GPT-4o miniClosedOpenAI · released Jul 18, 2024 · 13 days oldIntelligence 6.7$0.26 /M
- 10Llama 3.1 Instruct 70BOpen weightsMeta · released Jul 23, 2024 · 8 days oldIntelligence 6.6$0.56 /M
2 years later
What has happened since
450 models have come out since. Today's best is Claude Opus 5.5 at 57.6, 48.9 points higher.
Gemma 4 E4B scores at least as well today for $0.04 per million tokens, against $30 for Claude 3 Opus back then.
Grok 3 mini Reasoning matched it for $0.35 per million tokens on Feb 19, 2025.
DeepSeek R1 (Jan '25) from DeepSeek matched it on Jan 20, 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
50 models were out on Jul 31, 2024
Meta, scoring 7.3. #3 overall, 1.4 points behind.
Meta, scoring 6.9. #6 overall at $0.04, 750× cheaper than #1.
- Mistral Large 2 (Jul '24)6.8
- Llama 3.1 Instruct 405B7.3
- Llama 3.1 Instruct 8B6.9
- Llama 3.1 Instruct 70B6.6
- GPT-4o mini6.7
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.