summarization benchmark
SQuALITY
SQuALITY (Summarization-format QUestion Answering with Long Input Texts, Yes!) is a long-document summarization dataset built by hiring highly-qualified contractors to read public-domain short stories (3000-6000 words) and write original summaries from scratch. Each document has five summaries: one overview and four question-focused summaries. Designed to address limitations in existing summarization datasets by providing high-quality, faithful summaries.
Updated Aug 7, 2026
Published models5
Registry coverage5
MetricScore
EvidenceB
SQuALITY leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | MI | 24.3% | 100.0% | 5 | C | |
| 2 | MI | 24.1% | 75.0% | 5 | C | |
| 3 | AM Nova ProAmazon | 19.8% | 50.0% | 5 | C | |
| 4 | AMNova LiteAmazon | 19.2% | 25.0% | 5 | C | |
| 5 | AMNova MicroAmazon | 18.8% | 0.0% | 5 | C |
Score distribution
Top published rows on the benchmark's original scale.
SQuALITY highlights
The top published results on this benchmark's own scale.
What is SQuALITY?
Definition and scoring fields from the benchmark registry.
SQuALITY (Summarization-format QUestion Answering with Long Input Texts, Yes!) is a long-document summarization dataset built by hiring highly-qualified contractors to read public-domain short stories (3000-6000 words) and write original summaries from scratch. Each document has five summaries: one overview and four question-focused summaries. Designed to address limitations in existing summarization datasets by providing high-quality, faithful summaries.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- SQuALITY
- Modality
- text
- Primary category
- summarization
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- squality|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about SQuALITY.
Which model scores highest on SQuALITY?
Phi-3.5-mini-instruct is currently ranked first with 24.3%.
What does SQuALITY measure?
SQuALITY (Summarization-format QUestion Answering with Long Input Texts, Yes!) is a long-document summarization dataset built by hiring highly-qualified contractors to read public-domain short stories (3000-6000 words) and write original summaries from scratch. Each document has five summaries: one overview and four question-focused summaries. Designed to address limitations in existing summarization datasets by providing high-quality, faithful summaries.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
5 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.