reasoning benchmark
WideSearch Leaderboard
WideSearch is a reasoning benchmark for evaluating broad, parallel search across multiple sources.
Updated Oct 23, 2026
Models15
Model coverage15
MetricScore
EvidenceB
WideSearch Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelTE | Score83.90% | Percentile100.00% | Participants15 | EvidenceC | Evaluated |
| Rank02 | ModelAC | Score81.90% | Percentile92.86% | Participants15 | EvidenceC | Evaluated |
| Rank03 | ModelSA | Score81.90% | Percentile85.71% | Participants15 | EvidenceC | Evaluated |
| Rank04 | ModelMA | Score80.80% | Percentile78.57% | Participants15 | EvidenceC | Evaluated |
| Rank05 | ModelMA | Score79.00% | Percentile71.43% | Participants15 | EvidenceC | Evaluated |
| Rank06 | ModelTE | Score76.40% | Percentile64.29% | Participants15 | EvidenceC | Evaluated |
| Rank07 | ModelAC | Score74.30% | Percentile57.14% | Participants15 | EvidenceC | Evaluated |
| Rank08 | ModelAC | Score74.00% | Percentile50.00% | Participants15 | EvidenceC | Evaluated |
| Rank09 | ModelIN | Score73.60% | Percentile42.86% | Participants15 | EvidenceC | Evaluated |
| Rank10 | ModelBY | Score63.80% | Percentile35.71% | Participants15 | EvidenceC | Evaluated |
| Rank11 | ModelAC | Score61.10% | Percentile28.57% | Participants15 | EvidenceC | Evaluated |
| Rank12 | ModelAC | Score60.50% | Percentile21.43% | Participants15 | EvidenceC | Evaluated |
| Rank13 | ModelAC | Score60.10% | Percentile14.29% | Participants15 | EvidenceC | Evaluated |
| Rank14 | ModelAC | Score57.90% | Percentile7.14% | Participants15 | EvidenceC | Evaluated |
| Rank15 | ModelAC | Score57.10% | Percentile0.00% | Participants15 | EvidenceC | Evaluated |
WideSearch Highlights
The leading models and scores on this benchmark.
WideSearch Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for WideSearch
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is WideSearch?
What WideSearch measures and how its scores work.
WideSearch is an agentic search benchmark.
It measures models' ability to perform broad, parallel search operations across multiple sources, including wide-coverage information retrieval and synthesis.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- WideSearch
- 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 WideSearch.
Which model scores highest on WideSearch?
Hy4 preview is currently ranked first with 83.90%.
What are the top three models on WideSearch?
The current leaders are Hy4 preview (83.90%), Qwen3.8 Max (81.90%), and Atria Dawn Preview (81.90%).
Which WideSearch model has the lowest official input price?
Qwen3.5-35B-A3B has the lowest matched official input price at $0.25 input / $2.0 output per 1M tokens.
Which models are fastest among WideSearch results?
No matched runtime record is currently available.
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.