Which model leads Agent's Last Exam (Pass rate)?

Q: Which model leads Agent's Last Exam (Pass rate)?

Gemini 4 Argon tops the board at 39.5

Agent's Last Exam (Pass rate) Leaderboard — Model Scores

Pro Mode — raw scores, one benchmark under the lensofficial card harness

The field · 7 measured

24.239.5

7 models measured, most on the official card harness. Gemini 4 Argon tops the board at 39.5.

7 measured·Lifecycle Updated Oct 8, 2026Score basis

#◆Vendor$/M in$/M out

Leader — ignites whiteField — fill cools with scoreIndependently verified only0 of 7Bars show 24.2–39.5 · ◆ solid = first-party or better · outlined = arm's-length

01Gemini 4 ArgonGoogle39.5$2.0/M$10.0/M$0.152

02Claude Opus 5.5Anthropic38.2$4.0/M$20.0/M$0.314

03GPT-6 AstraOpenAI34.2$10.0/M$50.0/M$0.877

04Claude Sonnet 5Anthropic33.3$2.0/M$10.0/M$0.180

05GPT-5.6 TerraOpenAI28$2.0/M$12.0/M$0.250

06Gemini 3.7 FlashGoogle26.3$0.75/M$3.8/M$0.086

07Gemini 3.6 FlashGoogle24.2$0.75/M$3.8/M$0.093

◆ solid = independently verified · aggregator attested · vendor attributed — outlined = cross-reference · source-attributed. Dimmed ◆ = confidence inferred. “+N alts” = alternate disclosures in the drawer with why-demoted. Prices per million tokens; $/point = input price ÷ score. Every row opens its model dossier.

Verification: 7 models scored · 0 independently verified · 4 vendor cross-reference · 3 vendor-reported · 0 with source disagreement. How these tiers are assigned