LifeSci Diamond (2026-11; 4 November 2026 edition) - Plain API - Partial Credit: leaderboard
Arca’s 4 November 2026 edition regrades 55 regulatory affairs tasks with a revised strict-pass rubric and one GPT-6.1 Sol judge at medium reasoning effort, including OpenAI contestants. The native benchmark version is 2026-11-03-gdp-v6. The historical answer batches include 102 answers refreshed November 4. Ten example tasks are public; 45 are held out and 11 answer keys lack expert review. Three plain APIs use the shared Vercel AI Gateway reference harness, Exa snippet search and up to 40 steps; effective provider defaults were not recorded. Native CSV mean partial-credit precision is retained.
Metric: Mean partial-credit score (0-100). Source: www.arca.inc. Saturation forecast: Around February 2027. 3 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6 Astra | 83.79 | #5 |
| 2 | Claude Opus 5.5 | 65.31 | #3 |
| 3 | Claude Opus 5 | 56.2 | #16 |
No result here: #6 GPT-6.1 Sol, #9 Gemini 4 Argon, #10 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=lifesci-diamond-2026-11-4-november-2026-edition-plain-api-partial-credit · How It Works · Data refreshed daily, snapshot 2026-11-06.