reasoning benchmark
DeepPlanning
DeepPlanning evaluates LLMs on complex multi-step planning tasks requiring long-horizon reasoning, goal decomposition, and strategic decision-making.
Updated Aug 7, 2026
Published models9
Registry coverage9
MetricScore
EvidenceB
DeepPlanning leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | AC | 62.3% | 100.0% | 9 | C | |
| 2 | AC | 41.5% | 87.5% | 9 | C | |
| 3 | AC | 34.3% | 75.0% | 9 | C | |
| 4 | AC | 25.9% | 62.5% | 9 | C | |
| 5 | AC | 24.1% | 50.0% | 9 | C | |
| 6 | AC | 22.8% | 37.5% | 9 | C | |
| 7 | AC | 22.6% | 25.0% | 9 | C | |
| 8 | AC | 18.0% | 12.5% | 9 | C | |
| 9 | AC | 17.6% | 0.0% | 9 | C |
Score distribution
Top published rows on the benchmark's original scale.
DeepPlanning highlights
The top published results on this benchmark's own scale.
What is DeepPlanning?
Definition and scoring fields from the benchmark registry.
DeepPlanning evaluates LLMs on complex multi-step planning tasks requiring long-horizon reasoning, goal decomposition, and strategic decision-making.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- DeepPlanning
- Modality
- text
- Primary category
- reasoning
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- deep-planning|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about DeepPlanning.
Which model scores highest on DeepPlanning?
Qwen3.7-Plus is currently ranked first with 62.3%.
What does DeepPlanning measure?
DeepPlanning evaluates LLMs on complex multi-step planning tasks requiring long-horizon reasoning, goal decomposition, and strategic decision-making.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
9 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.