Werdykt v1 - Category Routing: leaderboard
Werdykt v1 Category Routing 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 | Jebadiah 27B (Werdykt Run: native=Jebadiah%2027B; cfg=b4d6cf3fec1768ac) | 99.8 | |
| 2 | basal-omni-1.0-ultra (Werdykt Run: native=basal-omni-1.0-ultra; cfg=bb9754f68d2ef61a) | 99.4 | |
| 3 | deck-31B (Werdykt Run: native=deck-31B; cfg=6896e74abae00c63) | 99.4 | |
| 4 | Gemini 3.8 Flash (Werdykt Run: native=Gemini%203.8%20Flash; cfg=d6e25ef16f84a68f) | 99.2 | |
| 5 | Liquid d1-ultra (Werdykt Run: native=Liquid%20d1-ultra; cfg=4faf298d55c875bb) | 99.2 | |
| 6 | pplx-decider-v1.1-27B (Decisions API) (Werdykt Run: native=pplx-decider-v1.1-27b%20%28Decisions%20API%29; cfg=560a4b78e1b7f693) | 99.2 | |
| 7 | Eikos-27B-FP8 (v1.3) (Werdykt Run: native=Eikos-27B-FP8%20%28v1.3%29; cfg=b80bd06dcaa2f2dc) | 99.2 | |
| 8 | pplx-decider-v1-27B (Werdykt Run: native=pplx-decider-v1-27b; cfg=dd1d846fc0af2505) | 99.2 | |
| 9 | simple-jev · Qwen3.8-27B (featherless) (Werdykt Run: native=simple-jev%20%C2%B7%20Qwen3.8-27B%20%28featherless%29; cfg=68e523cf6b564737) | 99.2 | |
| 10 | Reflex 27B (Decision Index 0.2 build) (Werdykt Run: native=Reflex%2027B%20%28Decision%20Index%200.2%20build%29; cfg=9a3291ae2fa36dc1) | 99.2 |
Interactive version: theaggregate.ai/benchmark?slug=werdykt-v1-category-routing · How It Works · Data refreshed daily, snapshot 2026-11-05.