Which model leads FrontierMath (overall)?

Q: Which model leads FrontierMath (overall)?

GPT-5.6 Sol tops the board at 89

FrontierMath (overall) Leaderboard — Model Scores

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

The field · 16 measured

5.589

16 models measured, most on the official card harness. GPT-5.6 Sol tops the board at 89.

16 measured·Lifecycle Updated Sep 29, 2026Basis

#◆Vendor$/M in$/M out

Leader — ignites whiteField — fill cools with scoreIndependently verified only0 of 16Bars show 5.5–89 · ◆ solid = first-party or better · outlined = arm's-length

01GPT-5.6 SolOpenAI89$4.0/M$20.0/M$0.135

02GPT-5.6 TerraOpenAI84.9$2.0/M$12.0/M$0.082

03GPT-5.6 LunaOpenAI78.6$0.20/M$1.2/M$0.009

04GPT-5.4OpenAI47.6$2.5/M$15.0/M$0.184

05GPT-5.2OpenAI40.3$1.8/M$14.0/M$0.195

06GPT-5.5 ProOpenAI39.6$30.0/M$180.0/M$2.652

07GPT-5.5OpenAI35.4$5.0/M$30.0/M$0.494

08GPT-5.1 InstantOpenAI26.7———

09GPT-5.1OpenAI26.7$1.3/M$10.0/M$0.211

10GPT-5OpenAI26.3$1.3/M$10.0/M$0.214

11GPT-5 miniOpenAI22.1$0.25/M$2.0/M$0.051

12O3OpenAI15.8$2.0/M$8.0/M$0.316

13GPT-5 nanoOpenAI9.6$0.05/M$0.40/M$0.023

14O3 MiniOpenAI9.2$1.1/M$4.4/M$0.299

15MAI-Code-1-FlashMicrosoft6.3———

16O1OpenAI5.5$15.0/M$60.0/M$6.818

◆ 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: 16 models scored · 0 independently verified · 16 vendor-reported · 0 with source disagreement. How these tiers are assigned