reasoning benchmark
AlpacaEval 2.0
AlpacaEval 2.0 is a length-controlled automatic evaluator for instruction-following language models that uses GPT-4 Turbo to assess model responses against a baseline. It evaluates models on 805 diverse instruction-following tasks including creative writing, classification, programming, and general knowledge questions. The benchmark achieves 0.98 Spearman correlation with ChatBot Arena while being fast (< 3 minutes) and affordable (< $10 in OpenAI credits). It addresses length bias in automatic evaluation through length-controlled win-rates and uses weighted scoring based on response quality.
Updated Aug 7, 2026
Published models4
Registry coverage4
MetricScore
EvidenceB
AlpacaEval 2.0 leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | IB | 62.7% | 100.0% | 4 | C | |
| 2 | IB | 62.7% | 66.7% | 4 | C | |
| 3 | DE | 50.5% | 33.3% | 4 | C | |
| 4 | IB | 35.2% | 0.0% | 4 | C |
Score distribution
Top published rows on the benchmark's original scale.
AlpacaEval 2.0 highlights
The top published results on this benchmark's own scale.
What is AlpacaEval 2.0?
Definition and scoring fields from the benchmark registry.
AlpacaEval 2.0 is a length-controlled automatic evaluator for instruction-following language models that uses GPT-4 Turbo to assess model responses against a baseline. It evaluates models on 805 diverse instruction-following tasks including creative writing, classification, programming, and general knowledge questions. The benchmark achieves 0.98 Spearman correlation with ChatBot Arena while being fast (< 3 minutes) and affordable (< $10 in OpenAI credits). It addresses length bias in automatic evaluation through length-controlled win-rates and uses weighted scoring based on response quality.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- AlpacaEval 2.0
- Modality
- text
- Primary category
- reasoning
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- alpacaeval-2.0|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about AlpacaEval 2.0.
Which model scores highest on AlpacaEval 2.0?
Granite 3.3 8B Base is currently ranked first with 62.7%.
What does AlpacaEval 2.0 measure?
AlpacaEval 2.0 is a length-controlled automatic evaluator for instruction-following language models that uses GPT-4 Turbo to assess model responses against a baseline. It evaluates models on 805 diverse instruction-following tasks including creative writing, classification, programming, and general knowledge questions. The benchmark achieves 0.98 Spearman correlation with ChatBot Arena while being fast (< 3 minutes) and affordable (< $10 in OpenAI credits). It addresses length bias in automatic evaluation through length-controlled win-rates and uses weighted scoring based on response quality.
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.