math benchmark
AlignBench
AlignBench is a comprehensive multi-dimensional benchmark for evaluating Chinese alignment of Large Language Models. It contains 8 main categories: Fundamental Language Ability, Advanced Chinese Understanding, Open-ended Questions, Writing Ability, Logical Reasoning, Mathematics, Task-oriented Role Play, and Professional Knowledge. The benchmark includes 683 real-scenario rooted queries with human-verified references and uses a rule-calibrated multi-dimensional LLM-as-Judge approach with Chain-of-Thought for evaluation.
Updated Aug 7, 2026
Published models4
Registry coverage4
MetricScore
EvidenceB
AlignBench leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | AC | 81.6% | 100.0% | 4 | C | |
| 2 | DE | 80.4% | 66.7% | 4 | C | |
| 3 | AC | 73.3% | 33.3% | 4 | C | |
| 4 | AC | 72.1% | 0.0% | 4 | C |
Score distribution
Top published rows on the benchmark's original scale.
AlignBench highlights
The top published results on this benchmark's own scale.
What is AlignBench?
Definition and scoring fields from the benchmark registry.
AlignBench is a comprehensive multi-dimensional benchmark for evaluating Chinese alignment of Large Language Models. It contains 8 main categories: Fundamental Language Ability, Advanced Chinese Understanding, Open-ended Questions, Writing Ability, Logical Reasoning, Mathematics, Task-oriented Role Play, and Professional Knowledge. The benchmark includes 683 real-scenario rooted queries with human-verified references and uses a rule-calibrated multi-dimensional LLM-as-Judge approach with Chain-of-Thought for evaluation.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- AlignBench
- Modality
- text
- Primary category
- math
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- alignbench|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about AlignBench.
Which model scores highest on AlignBench?
Qwen2.5 72B Instruct is currently ranked first with 81.6%.
What does AlignBench measure?
AlignBench is a comprehensive multi-dimensional benchmark for evaluating Chinese alignment of Large Language Models. It contains 8 main categories: Fundamental Language Ability, Advanced Chinese Understanding, Open-ended Questions, Writing Ability, Logical Reasoning, Mathematics, Task-oriented Role Play, and Professional Knowledge. The benchmark includes 683 real-scenario rooted queries with human-verified references and uses a rule-calibrated multi-dimensional LLM-as-Judge approach with Chain-of-Thought for evaluation.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
4 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is preserved as source-native evidence but is not eligible for the current overall score.