Which model leads DROP (F1)?

Q: Which model leads DROP (F1)?

Gemini 2.0 Flash Exp tops the board at 89.3

DROP (F1) Leaderboard — Model Scores

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

The field · 12 measured

80.489.3

12 models measured, most on the official card harness. Gemini 2.0 Flash Exp tops the board at 89.3.

12 measured·Lifecycle Updated Oct 8, 2026Score basis

#◆Vendor$/M in$/M out

Leader — ignites whiteField — fill cools with scoreIndependently verified only0 of 12Bars show 80.4–89.3 · ◆ solid = first-party or better · outlined = arm's-length

01Gemini 2.0 Flash ExpGoogle89.33P———

02GPT-4oOpenAI89.23P$5.0/M$15.0/M$0.112

03Gemini 1.5 ProGoogle89.23P———

04DeepSeek-V3+1 altDeepSeek89$0.24/M$0.90/M$0.006

05Claude 3.5 SonnetAnthropic88.83P$3.0/M$15.0/M$0.101

06DeepSeek-V4-Pro-Base+4 altsDeepSeek88.7———

07DeepSeek-V4-Flash-Base+3 altsDeepSeek88.6———

08DeepSeek-V3.2-Base+2 altsDeepSeek88.2———

09MiniMax Text 01MiniMax87.83P———

10Llama 3.1 Instruct 405B+1 altMeta86———

11Qwen2.5 Instruct 72B+1 altAlibaba80.6$0.47/M$0.49/M$0.006

12DeepSeek-V2DeepSeek80.4———

◆ 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: 12 models scored · 0 independently verified · 6 vendor cross-reference · 6 vendor-reported · 0 with source disagreement. How these tiers are assigned