Werdykt v1 - Category Realdoc: leaderboard
Werdykt v1 Category Realdoc measures text-only decisions in Polish and English. The full test has 5,000 hidden questions in twenty balanced language/category groups of 250; category panels combine 500 questions and language panels have 2,500. Macro accuracy gives groups equal weight. Request errors and unparsed answers count incorrect. Native non-control systems retain their interface, engine, reasoning, legacy and source-independence flags; decision endpoints remain distinct from chat.
Metric: Accuracy (%). Source: basal.si5.pl. Saturation forecast: Rough model projection: around 2026. 121 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.8 Flash (Werdykt Run: native=Gemini%203.8%20Flash; cfg=d6e25ef16f84a68f) | 99.6 | |
| 2 | GPT-6 Luna (Werdykt Run: native=GPT-6%20Luna; cfg=b572761faa59abee) | 99.6 | |
| 3 | pplx-decider-v1.1-27B (Decisions API) (Werdykt Run: native=pplx-decider-v1.1-27b%20%28Decisions%20API%29; cfg=560a4b78e1b7f693) | 99.6 | |
| 4 | Mercury Decide (OpenRouter Decisions API) (Werdykt Run: native=Mercury%20Decide%20%28OpenRouter%20Decisions%20API%29; cfg=cf8e2aa76c100016) | 99.6 | |
| 5 | simple-jev · Qwen3.8-27B (featherless) (Werdykt Run: native=simple-jev%20%C2%B7%20Qwen3.8-27B%20%28featherless%29; cfg=68e523cf6b564737) | 99.6 | |
| 6 | Gemini 3.1 Pro (Preview) (Werdykt Run: native=Gemini%203.1%20Pro%20%28preview%29; cfg=e8ffbba3cfe39393) | 99.4 | |
| 7 | Jev 1.14 (Werdykt Run: native=Jev%201.14; cfg=b99826cc6e3eaa9a) | 99.4 | |
| 8 | pplx-decider-v1-27B (Werdykt Run: native=pplx-decider-v1-27b; cfg=dd1d846fc0af2505) | 99.4 | |
| 9 | Jev 1.13.0 (Werdykt Run: native=Jev%201.13.0; cfg=afcc43417b53c1dd) | 99.4 | |
| 10 | GPT-6.1 Sol (Werdykt Run: native=GPT-6.1%20Sol; cfg=9bc38d71de848a6b) | 99.2 |
Interactive version: theaggregate.ai/benchmark?slug=werdykt-v1-category-realdoc · How It Works · Data refreshed daily, snapshot 2026-11-05.