reasoning benchmark
DeepSearchQA
DeepSearchQA is a benchmark for evaluating deep search and question-answering capabilities, testing models' ability to perform multi-hop reasoning and information retrieval across complex knowledge domains.
Updated Aug 7, 2026
Published models8
Registry coverage8
MetricScore
EvidenceB
DeepSearchQA leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | MA | 95.0% | 100.0% | 8 | C | |
| 2 | AN | 93.1% | 85.7% | 8 | C | |
| 3 | AN | 91.3% | 71.4% | 8 | C | |
| 4 | TE | 91.0% | 57.1% | 8 | C | |
| 5 | XI | 86.7% | 42.9% | 8 | C | |
| 6 | MA | 83.0% | 28.6% | 8 | C | |
| 7 | MA | 77.1% | 14.3% | 8 | C | |
| 8 | ME | 74.8% | 0.0% | 8 | C |
Score distribution
Top published rows on the benchmark's original scale.
DeepSearchQA highlights
The top published results on this benchmark's own scale.
What is DeepSearchQA?
Definition and scoring fields from the benchmark registry.
DeepSearchQA is a benchmark for evaluating deep search and question-answering capabilities, testing models' ability to perform multi-hop reasoning and information retrieval across complex knowledge domains.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- DeepSearchQA
- Modality
- text
- Primary category
- reasoning
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- deepsearchqa|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about DeepSearchQA.
Which model scores highest on DeepSearchQA?
Kimi K3 is currently ranked first with 95.0%.
What does DeepSearchQA measure?
DeepSearchQA is a benchmark for evaluating deep search and question-answering capabilities, testing models' ability to perform multi-hop reasoning and information retrieval across complex knowledge domains.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
8 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.