Knowledge Bases and Evaluation Organizations: SuperCLUE
Knowledge Bases and Evaluation Organizations · SuperCLUE. Monthly leaderboard of Chinese LLMs on newly written questions. This source has 4 entries admitted under this category's review policy; each links directly to the original.
Authority tiers in this category: A2–B. Admission policy: All core resources are included. Each secondary source is admitted according to its own database, methodology, versions and leaderboard structure; there is no cross-organization scoring scheme.
SuperCLUE general leaderboard
Updates a leaderboard of Chinese models across reasoning, science, code and agentic programming, instruction following, hallucination control and task planning.
Evidence use: Supports Chinese models' dimension scores and relative positions under the question set and weights of that period.
Limitations: Questions and weights are adjusted over time. Scores from different periods cannot be compared directly.
Related concepts: Model Evaluation and Benchmarks
Link last checked: 2026-09-24
SuperCLUE large model evaluation system
Outlines evaluation areas such as general, multimodal, agent and industry models, with links to the official leaderboards.
Evidence use: Supports describing SuperCLUE's evaluation scope and product structure.
Limitations: The overview page does not replace the methodology of each leaderboard. Some services are commercial evaluations.
Related concepts: Model Evaluation and Benchmarks · AI Agent
Link last checked: 2026-09-24
SuperCLUE Agent Benchmark 2025H1
Evaluates agents' planning, tool use and task completion in Chinese-language scenarios.
Evidence use: Supports versioned comprehensive comparison of Chinese agents.
Limitations: Leaderboard question sets and weights change between versions. Scores from different versions cannot be compared directly.
Related concepts: AI Agent · Model Evaluation and Benchmarks
Link last checked: 2026-09-24
AgentCLUE-Mobile
Evaluates agents' perception, planning and execution on real tasks in mobile apps.
Evidence use: Supports comparing the capabilities of mobile GUI agents.
Limitations: Leaderboard question sets and weights change between versions. Scores from different versions cannot be compared directly.
Related concepts: AI Agent · Model Evaluation and Benchmarks
Link last checked: 2026-09-24