productivity benchmark
DECK-Bench
DECK-Bench is Moonshot AI's internal evaluation of agents on creating and reasoning over presentation-style knowledge-work artifacts.
Updated Aug 11, 2026
Models1
Model coverage1
MetricScore
EvidenceB
DECK-Bench Ranking
Higher score ranks better on this benchmark.
| 01 | MA | 73.5% | 100.0% | 1 | C |
DECK-Bench Highlights
The leading models and scores on this benchmark.
What is DECK-Bench?
What DECK-Bench measures and how its scores work.
DECK-Bench is Moonshot AI's internal evaluation of agents on creating and reasoning over presentation-style knowledge-work artifacts.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- DECK-Bench
- Modality
- text
- Primary category
- productivity
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- deck-bench|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about DECK-Bench.
Which model scores highest on DECK-Bench?
Kimi K3 is currently ranked first with 73.5%.
What does DECK-Bench measure?
DECK-Bench is Moonshot AI's internal evaluation of agents on creating and reasoning over presentation-style knowledge-work artifacts.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
1 model results are currently shown.
Does this benchmark affect the overall score?
No. This benchmark is shown for reference but does not contribute to the overall score.