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

BenchmarkScoreMetricPercentile
LLM2014 Logic 2025-1240.88Median Score62
LLM2014 Logic 2026-0138.67Median Score61.4
CLBench - Procedural Task Execution19.4Solving Rate (%)54.3
CLBench - Rule System Application17.3Solving Rate (%)45.7
LLM2014 Logic 2026-0232.03Median Score42.2
CLBench - Domain Knowledge Reasoning18Solving Rate (%)41.4
CLBench17.2Solving Rate (%)40
LLM2014 Logic 2026-0333.22Median Score39
CLBench Life - Fragmented Information & Revisions9.4Solving Rate (%)25
CLBench - Empirical Discovery & Simulation8.9Solving Rate (%)18.6
CLBench Life - Behavioral Records & Activity Trails6.4Solving Rate (%)17.9
CLBench Life8.4Solving Rate (%)16.1

Interactive version: theaggregate.ai/model?slug=hy-2-0-thinking · How It Works · Data refreshed daily, snapshot 2026-09-19.