US: AI model benchmarks
AI models developed in US. Rank among countries: #1. Strongest model: Claude Mythos 5.1 (Rating 2019, #1 overall, private). Best available: Claude Opus 5.5 (Rating 1986, #3 overall). Latest release: Beam 1.0 (Oct 16, 2026).
Current models, released in the last 18 months
| Rank | Model | Provider | Rating | Released | Access |
|---|---|---|---|---|---|
| 1 | Claude Mythos 5.1 | Anthropic | 2019 ± 29 | Sep 1, 2026 | Private |
| 2 | Claude Mythos 5 | Anthropic | 2009 ± 13 | Jun 9, 2026 | Private |
| 3 | Claude Opus 5.5 | Anthropic | 1986 ± 5 | Sep 22, 2026 | API |
| 4 | Claude Mythos Preview | Anthropic | 1984 ± 12 | Apr 7, 2026 | Private |
| 5 | GPT-6 Astra Pro | OpenAI | 1984 ± 21 | Sep 3, 2026 | API |
| 6 | GPT-6 Astra | OpenAI | 1983 ± 4 | Sep 3, 2026 | API |
| 7 | GPT-6.1 Sol | OpenAI | 1969 ± 6 | Sep 29, 2026 | API |
| 8 | Gemini 4 Argon | 1967 ± 7 | Sep 30, 2026 | API | |
| 9 | Claude Fable 5.1 | Anthropic | 1958 ± 5 | Sep 1, 2026 | API |
| 10 | Muse Wonder Full | Meta | 1939 ± 20 | Oct 12, 2026 | Interface |
| 11 | Claude Sonnet 5.5 | Anthropic | 1934 ± 7 | Sep 28, 2026 | API |
| 12 | Claude Opus 5 | Anthropic | 1931 ± 3 | Jul 24, 2026 | API |
| 13 | GPT-6.1 Pro Sol | OpenAI | 1924 ± 29 | Sep 29, 2026 | API |
| 14 | GPT-6 Sol | OpenAI | 1910 ± 6 | Sep 22, 2026 | API |
| 15 | Muse Wonder | Meta | 1908 ± 9 | Oct 12, 2026 | API |
| 16 | Claude Fable 5 | Anthropic | 1907 ± 4 | Jun 9, 2026 | API |
| 17 | GPT-5.5 Pro | OpenAI | 1899 ± 13 | Apr 23, 2026 | API |
| 18 | GPT-5.6 Pro Sol | OpenAI | 1890 ± 14 | Jul 9, 2026 | API |
| 19 | GPT-5.6 Sol | OpenAI | 1889 ± 3 | Jul 9, 2026 | API |
| 21 | Gemini 3.8 Flash | 1873 ± 5 | Sep 2, 2026 | API |
Interactive version: theaggregate.ai/country?slug=us · How It Works · Data refreshed daily, snapshot 2026-10-20.