Which model leads AndroidWorld?

Q: Which model leads AndroidWorld?

Qwen3.8 Max Preview tops the board at 85.3

AndroidWorld Leaderboard — Model Scores

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

The field · 9 measured

2285.3

9 models measured, most on the official card harness. Qwen3.8 Max Preview tops the board at 85.3.

9 measured·Lifecycle Updated Oct 17, 2026Score basis

#◆Vendor$/M in$/M out

Leader — ignites whiteField — fill cools with scoreIndependently verified only0 of 9Bars show 22–85.3 · ◆ solid = first-party or better · outlined = arm's-length

01Qwen3.8 Max PreviewAlibaba85.3$3.3/M$9.9/M$0.077

02Qwen3.8 Flash Next+1 altAlibaba84.5$0.15/M$0.47/M$0.004

03Qwen3.8 27B+1 altAlibaba81.9$0.50/M$3.0/M$0.021

04Qwen3.7 Plus Preview+1 altAlibaba81$0.40/M$1.6/M$0.012

05GLM 5V TurboZ.ai75.7$1.2/M$4.0/M$0.034

06Seed 1.8ByteDance70.7$0.25/M$2.0/M$0.016

07Qwen3.6 27BAlibaba70.3$0.60/M$3.6/M$0.030

08Claude Opus 4.6Anthropic62$5.0/M$25.0/M$0.242

09Qwen2.5 VL 32B InstructAlibaba22———

◆ 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: 9 models scored · 0 independently verified · 1 vendor cross-reference · 8 vendor-reported · 0 with source disagreement. How these tiers are assigned