Which model leads ProgramBench (Almost Solved)?

Q: Which model leads ProgramBench (Almost Solved)?

GPT-5.6 Sol tops the board at 23

ProgramBench (Almost Solved) Leaderboard — Model Scores

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

The field · 6 measured

9.523

6 models measured, most on the third-party eval harness. GPT-5.6 Sol tops the board at 23.

6 measured·Lifecycle Updated Oct 17, 2026Score basis

#◆Vendor$/M in$/M out

Leader — ignites whiteField — fill cools with scoreIndependently verified only0 of 6Bars show 9.5–23 · ◆ solid = first-party or better · outlined = arm's-length

01GPT-5.6 Sol+1 altOpenAI23$4.0/M$20.0/M$0.522

02GLM-5.3+1 altZ.ai19$1.4/M$4.4/M$0.153

03Kimi K3+1 altMoonshot17.5$0.30/M$12.3/M$0.360

04Claude Opus 4.8+1 altAnthropic15.5$5.0/M$25.0/M$0.968

05Qwen3.8 Max Preview+1 altAlibaba10.5$3.3/M$9.9/M$0.629

06GLM-5.2+1 altZ.ai9.5$1.4/M$4.4/M$0.305

◆ 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