Which model leads LiveCodeBenchV6 no tools?

LiveCodeBenchV6 no tools Leaderboard — Model Scores

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

The field · 10 measured

54.9287

10 models measured, most on the official card harness. GPT-5 tops the board at 87.

10 measured·Lifecycle Updated Oct 9, 2026Score basis

#◆Vendor$/M in$/M out

Leader — ignites whiteField — fill cools with scoreIndependently verified only0 of 10Bars show 54.92–87 · ◆ solid = first-party or better · outlined = arm's-length

01GPT-5OpenAI873P$1.3/M$10.0/M$0.065

02Kimi K2 ThinkingMoonshot83.1$0.60/M$2.5/M$0.019

03DeepSeek-V3.2DeepSeek74.13P$0.28/M$0.42/M$0.005

04Qwen3.5 9B+1 altAlibaba69.86$0.14/M$0.20/M$0.002

05Claude Sonnet 4.5Anthropic643P$3.0/M$15.0/M$0.141

06Gemma 4 E4B+1 altGoogle63.77$0.02/M$0.10/M$0.001

07Qwen3.5 4B+1 altAlibaba60.85$0.03/M$0.15/M$0.001

08LFM2.5-2.6B+1 altLiquid AI59.41———

09Kimi K2 InstructMoonshot56.13P$0.57/M$2.3/M$0.026

10Gemma 4 E2B+1 altGoogle54.92———

◆ 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: 10 models scored · 0 independently verified · 3 vendor cross-reference · 7 vendor-reported · 0 with source disagreement. How these tiers are assigned