Which model leads HealthBench length-adjusted?

Q: Which model leads HealthBench length-adjusted?

Claude Opus 5.5 tops the board at 60.6

HealthBench length-adjusted Leaderboard

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

The field · 14 measured

50.960.6

14 models measured, most on the official card harness. Claude Opus 5.5 tops the board at 60.6.

14 measured·Lifecycle Updated Oct 8, 2026Score basis

#◆Vendor$/M in$/M out

Leader — ignites whiteField — fill cools with scoreIndependently verified only0 of 14Bars show 50.9–60.6 · ◆ solid = first-party or better · outlined = arm's-length

01Claude Opus 5.5Anthropic60.6$4.0/M$20.0/M$0.198

02Claude Fable 5.1Anthropic60$10.0/M$50.0/M$0.500

03GPT-6 Astra+1 altOpenAI58.3$10.0/M$50.0/M$0.515

04Claude Opus 5Anthropic57.8$5.0/M$25.0/M$0.260

05GPT-5OpenAI57.7$1.3/M$10.0/M$0.097

06GPT-5.6 Terra+1 altOpenAI57$2.0/M$12.0/M$0.123

07GPT-5.6 Sol+1 altOpenAI57$4.0/M$20.0/M$0.211

08GPT-5.2OpenAI56.8$1.8/M$14.0/M$0.139

09GPT-5.5+1 altOpenAI56.5$5.0/M$30.0/M$0.310

10GPT-5.6 Luna+2 altsOpenAI55.8$0.20/M$1.2/M$0.013

11GPT-6 LunaOpenAI54.5$0.10/M$0.50/M$0.006

12GPT-5.4OpenAI54$2.5/M$15.0/M$0.162

13GPT-6 SolOpenAI53.2$2.0/M$10.0/M$0.113

14GPT-5.1OpenAI50.9$1.3/M$10.0/M$0.111

◆ 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: 14 models scored · 0 independently verified · 14 vendor-reported · 0 with source disagreement. How these tiers are assigned