benchmark
.
space
benchmarks
rankings
compare
voices
transcripts
articles
USAMO 2026 leaderboard
USAMO 2026
3 models tested · Updated 2026-04-07 · Verified sources only
Claude Mythos Preview
leads at
97.6%
1
Claude Mythos Preview
Anthropic ·
Blog/Anthropic
· 2026-04-07
Near-perfect score on proof-based USAMO. Opus 4.6 scored 42.3%, GPT-5.4 95.2%. Massive capability leap.
97.6%
2
GPT-5.4
OpenAI ·
Blog/Anthropic
· 2026-04-07
Saturated USAMO proofs. Only significant error on Problem 5 (invalid counterexample). Year-over-year jump from disastrous 2025 results.
95.2%
3
Claude Opus 4.6
Anthropic ·
Blog/Anthropic
· 2026-04-07
Anthropic reported score. Dramatically behind Mythos (97.6%) and GPT-5.4 (95.2%).
42.3%