reasoning benchmark
DeepSearchQA Leaderboard
DeepSearchQA is a benchmark for evaluating deep search and question-answering capabilities across complex knowledge domains.
Updated Oct 23, 2026
Models13
Model coverage13
MetricScore
EvidenceB
DeepSearchQA Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelSA | Score96.00% | Percentile100.00% | Participants13 | EvidenceC | Evaluated |
| Rank02 | ModelMA | Score95.00% | Percentile91.67% | Participants13 | EvidenceC | Evaluated |
| Rank03 | ModelAN | Score93.10% | Percentile83.33% | Participants13 | EvidenceC | Evaluated |
| Rank04 | ModelAN | Score91.30% | Percentile75.00% | Participants13 | EvidenceC | Evaluated |
| Rank05 | ModelTE | Score91.00% | Percentile66.67% | Participants13 | EvidenceC | Evaluated |
| Rank06 | ModelME | Score89.40% | Percentile58.33% | Participants13 | EvidenceC | Evaluated |
| Rank07 | ModelXI | Score86.70% | Percentile50.00% | Participants13 | EvidenceC | Evaluated |
| Rank08 | ModelMA | Score83.00% | Percentile41.67% | Participants13 | EvidenceC | Evaluated |
| Rank09 | ModelRA | Score80.50% | Percentile33.33% | Participants13 | EvidenceC | Evaluated |
| Rank10 | ModelMA | Score77.10% | Percentile25.00% | Participants13 | EvidenceC | Evaluated |
| Rank11 | ModelME | Score74.80% | Percentile16.67% | Participants13 | EvidenceC | Evaluated |
| Rank12 | ModelME | Score74.60% | Percentile8.33% | Participants13 | EvidenceC | Evaluated |
| Rank13 | ModelIN | Score71.33% | Percentile0.00% | Participants13 | EvidenceC | Evaluated |
DeepSearchQA Highlights
The leading models and scores on this benchmark.
DeepSearchQA Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for DeepSearchQA
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is DeepSearchQA?
What DeepSearchQA measures and how its scores work.
DeepSearchQA is a reasoning benchmark for evaluating deep search and question-answering capabilities.
It measures models' ability to perform multi-hop reasoning and information retrieval across complex knowledge domains, using the Score metric as a ratio.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- DeepSearchQA
- Modality
- text
- Primary category
- reasoning
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- overall
LLM Stats. Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about DeepSearchQA.
Which model scores highest on DeepSearchQA?
Atria Dawn Preview is currently ranked first with 96.00%.
What are the top three models on DeepSearchQA?
The current leaders are Atria Dawn Preview (96.00%), Kimi K3 (95.00%), and Claude Opus 4.8 (93.10%).
Which DeepSearchQA model has the lowest official input price?
Mercury 3 has the lowest matched official input price at $0.10 input / $0.35 output per 1M tokens.
Which models are fastest among DeepSearchQA results?
The fastest matched records are Mercury 3 (1,068.00 tok/s via Inception), Claude Opus 4.8 (42.00 tok/s via Vertex AI), and Claude Opus 4.6 (42.00 tok/s via Anthropic).
Does the highest score result prove overall model quality?
No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.