HY 2.0 (Thinking): benchmark results
Provider: Other. Released 2025-11-09. Access: API.
Unified ELO 1641 ± 1, rank #388 of 3078 rated models, from 13 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| LLM2014 Logic 2025-12 | 40.88 | Median Score | 62 |
| LLM2014 Logic 2026-01 | 38.67 | Median Score | 61.4 |
| CLBench - Procedural Task Execution | 19.4 | Solving Rate (%) | 54.3 |
| CLBench - Rule System Application | 17.3 | Solving Rate (%) | 45.7 |
| LLM2014 Logic 2026-02 | 32.03 | Median Score | 42.2 |
| CLBench - Domain Knowledge Reasoning | 18 | Solving Rate (%) | 41.4 |
| CLBench | 17.2 | Solving Rate (%) | 40 |
| LLM2014 Logic 2026-03 | 33.22 | Median Score | 39 |
| CLBench Life - Fragmented Information & Revisions | 9.4 | Solving Rate (%) | 25 |
| CLBench - Empirical Discovery & Simulation | 8.9 | Solving Rate (%) | 18.6 |
| CLBench Life - Behavioral Records & Activity Trails | 6.4 | Solving Rate (%) | 17.9 |
| CLBench Life | 8.4 | Solving Rate (%) | 16.1 |
Interactive version: theaggregate.ai/model?slug=hy-2-0-thinking · How It Works · Data refreshed daily, snapshot 2026-09-19.