Which model leads Multilingual MMLU?

Multilingual MMLU Leaderboard — Model Scores

Pro Mode — raw scores, one benchmark under the lensthird-party eval harness

The field · 8 measured

49.380.7

8 models measured, most on the third-party eval harness. O3 Mini tops the board at 80.7.

8 measured·Lifecycle Updated Oct 8, 2026Score basis

#◆Vendor$/M in$/M out

Leader — ignites whiteField — fill cools with scoreIndependently verified only0 of 8Bars show 49.3–80.7 · ◆ solid = first-party or better · outlined = arm's-length

01O3 MiniOpenAI80.7VENDOR$1.1/M$4.4/M$0.034

02Qwen3 14B Base+3 altsAlibaba75.4———

03Ministral 3 14B+3 altsMistral74.2VENDOR$0.20/M$0.20/M$0.003

04Ministral 3 8B+2 altsMistral70.6$0.15/M$0.15/M$0.002

05Qwen3 8B Base+3 altsAlibaba70———

06Qwen3 4B Base+3 altsAlibaba67.7———

07Ministral 3 3B+2 altsMistral65.2$0.10/M$0.10/M$0.002

08Phi 4 Mini InstructMicrosoft49.3VENDOR———

◆ 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: 8 models scored · 0 independently verified · 3 vendor cross-reference · 5 vendor-reported · 0 with source disagreement. How these tiers are assigned