Which model leads OCRBench (Acc)?

Q: Which model leads OCRBench (Acc)?

Qwen2.5 VL 72B tops the board at 885

OCRBench (Acc) Leaderboard — 7 Models Ranked

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

The field · 7 measured

702885

7 models measured, most on the third-party eval harness. Qwen2.5 VL 72B tops the board at 885.

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 702–885 · ◆ solid = first-party or better · outlined = arm's-length

01Qwen2.5 VL 72BAlibaba885———

02Kimi VL A3B Instruct+1 altMoonshot867VENDOR———

03Kimi VL A3B Thinking+1 altMoonshot864———

04GPT-4o+1 altOpenAI815$5.0/M$15.0/M$0.012

05GPT-4o miniOpenAI785$0.15/M$0.60/M$0.000

06Gemma 3 27BGoogle753$0.08/M$0.16/M$0.000

07Gemma 3 12B+1 altGoogle702$0.05/M$0.15/M$0.000

◆ 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 · 5 vendor cross-reference · 2 vendor-reported · 0 with source disagreement. How these tiers are assigned