Which model leads PostTrainBench?

Q: Which model leads PostTrainBench?

Claude Opus 5.5 tops the board at 49.3

PostTrainBench Leaderboard — 16 Models Ranked

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

The field · 16 measured

16.549.3

16 models measured, most on the official card harness. Claude Opus 5.5 tops the board at 49.3.

16 measured·1 new this week·Lifecycle Updated Oct 8, 2026Score basis

#◆Vendor$/M in$/M out

Leader — ignites whiteField — fill cools with scoreIndependently verified only0 of 16Bars show 16.5–49.3 · ◆ solid = first-party or better · outlined = arm's-length

01Claude Opus 5.5Anthropic49.3$4.0/M$20.0/M$0.243

02Gemini 4 ArgonNEW+2 altsGoogle45.3$2.0/M$10.0/M$0.132

03GPT-6 AstraOpenAI44.3$10.0/M$50.0/M$0.677

04Claude Fable 5Anthropic41.43P$10.0/M$50.0/M$0.725

05Claude Fable 5.1Anthropic40.2$10.0/M$50.0/M$0.746

06GLM-5.3+2 altsZ.ai39.8$1.4/M$4.4/M$0.073

07Claude Opus 4.8+4 altsAnthropic37.2$5.0/M$25.0/M$0.403

08MiniMax M3MiniMax37.1$0.28/M$1.1/M$0.019

09Kimi K3+3 altsMoonshot36.6$0.58/M$12.3/M$0.176

10GPT-5.6 Sol+2 altsOpenAI36.23P$4.0/M$20.0/M$0.331

11Hy4Tencent35.6$0.83/M$2.5/M$0.047

12GLM-5.2+6 altsZ.ai34.3$1.4/M$4.4/M$0.085

13GPT-5.5+2 altsOpenAI28.4$5.0/M$30.0/M$0.616

14Gemini 3.1 Pro+1 altGoogle21.6$2.0/M$12.0/M$0.324

15GLM-5.1+1 altZ.ai20.1$1.4/M$4.4/M$0.144

16Seed 2.1 Pro PreviewByteDance16.5———

◆ 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: 16 models scored · 0 independently verified · 8 vendor cross-reference · 8 vendor-reported · 0 with source disagreement. How these tiers are assigned