Which model leads LiveCodeBench (24/8~25/5)?

Q: Which model leads LiveCodeBench (24/8~25/5)?

Gemini 2.5 Pro tops the board at 77.1

LiveCodeBench (24/8~25/5) Leaderboard — Model Scores

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

The field · 7 measured

55.977.1

7 models measured, most on the third-party eval harness. Gemini 2.5 Pro tops the board at 77.1.

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 55.9–77.1 · ◆ solid = first-party or better · outlined = arm's-length

01Gemini 2.5 Pro+1 altGoogle77.1$1.3/M$10.0/M$0.073

02O3+1 altOpenAI75.8$2.0/M$8.0/M$0.066

03Qwen3 235B A22B+1 altAlibaba65.9$0.70/M$2.8/M$0.027

04MiniMax M1 80K+1 altMiniMax65$0.55/M$2.2/M$0.021

05MiniMax M1 40K+1 altMiniMax62.3———

06Claude Opus 4+1 altAnthropic56.6$15.0/M$75.0/M$0.795

07DeepSeek-R1+1 altDeepSeek55.9$2.0/M$4.0/M$0.054

◆ 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