Which model leads OctoCodingbench?

Q: Which model leads OctoCodingbench?

Claude Opus 4.5 tops the board at 36.2

OctoCodingbench Leaderboard — Model Scores

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

The field · 6 measured

13.336.2

6 models measured, most on the third-party eval harness. Claude Opus 4.5 tops the board at 36.2.

6 measured·Lifecycle Updated Oct 8, 2026Score basis

#◆Vendor$/M in$/M out

Leader — ignites whiteField — fill cools with scoreIndependently verified only0 of 6Bars show 13.3–36.2 · ◆ solid = first-party or better · outlined = arm's-length

01Claude Opus 4.5Anthropic36.2$5.0/M$25.0/M$0.414

02MiniMax M2.1+1 altMiniMax26.1VENDOR$0.30/M$1.2/M$0.029

03DeepSeek-V3.2DeepSeek26$0.28/M$0.42/M$0.013

04Gemini 3 ProGoogle22.9$2.0/M$12.0/M$0.306

05Claude Sonnet 4.5Anthropic22.8$3.0/M$15.0/M$0.395

06MiniMax M2MiniMax13.3$0.30/M$1.2/M$0.056

◆ 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: 6 models scored · 0 independently verified · 4 vendor cross-reference · 2 vendor-reported · 0 with source disagreement. How these tiers are assigned