Which model leads Agents' Last Exam (ALE-CLI)?

Q: Which model leads Agents' Last Exam (ALE-CLI)?

GPT-5.6 Sol tops the board at 28.6

Agents' Last Exam (ALE-CLI) Leaderboard — Model Scores

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

The field · 7 measured

23.828.6

7 models measured, most on the third-party eval harness. GPT-5.6 Sol tops the board at 28.6.

7 measured·Lifecycle Updated Oct 17, 2026Score basis

#◆Vendor$/M in$/M out

Leader — ignites whiteField — fill cools with scoreIndependently verified only0 of 7Bars show 23.8–28.6 · ◆ solid = first-party or better · outlined = arm's-length

01GPT-5.6 Sol+1 altOpenAI28.6$4.0/M$20.0/M$0.420

02GLM-5.3+1 altZ.ai28.5$1.4/M$4.4/M$0.102

03Kimi K3+1 altMoonshot27.6$0.30/M$11.9/M$0.222

04Qwen3.8 Max Preview+1 altAlibaba27$3.3/M$9.9/M$0.244

05Claude Opus 4.8+1 altAnthropic25.7$5.0/M$25.0/M$0.584

06DeepSeek-V4-Pro 0813+1 altDeepSeek25.7$0.66/M$2.0/M$0.051

07GLM-5.2+1 altZ.ai23.8$1.4/M$4.4/M$0.122

◆ 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