Agentic benchmark
DeepSearchQA leaderboard
Every model the catalog carries a published DeepSearchQA value for, ranked by that value.
CategoryAgentic
MeasureSearch / open / find browser-agent evaluation
TasksAgentic browsing and list-answer questions
DifficultyAgentic web research
Published byMuse Spark Eval Methodology
An agentic browsing benchmark where models search the web, gather evidence, and answer list-style questions using browser tools.
DeepSearchQA ranking
18 models with a published DeepSearchQA value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.
| Rank | Model | Provider | Search / open / find browser-agent evaluation |
|---|---|---|---|
| 1 |  Atria Dawn Preview | Shanghai Artificial Intelligence Laboratory | 96.0 |
| 2 |  Claude Opus 5 | Anthropic | 95.0 |
| 2 |  Kimi K3 | Moonshot AI | 95.0 |
| 4 | Claude Opus 4.8 | Anthropic | 93.1 |
| 5 |  Step 3.7 Flash | StepFun | 92.8 |
| 6 | Kimi K2.6 | Moonshot AI | 92.5 |
| 7 | APApodex 1.1 | Apodex | 92.4 |
| 8 | DSdots3-note Preview | Dots Studio | 92.1 |
| 9 |  Muse Spark 1.3 | Meta | 89.4 |
| 10 | Muse Spark 1.1 | Meta | 84.9 |
| 11 | Kimi K2.5 | Moonshot AI | 77.1 |
| 12 | Muse Spark | Meta | 74.8 |
| 13 | Muse Glimmer 30B | Meta | 74.6 |
| 14 | Claude Opus 4.6 | Anthropic | 73.7 |
| 15 |  GPT-5.4 | OpenAI | 73.6 |
| 16 |  Gemini 3.1 Pro | 69.7 | |
| 17 |  Grok 4.20 | xAI | 62.8 |
| 18 |  Mercury 2.5 | Inception | 34.0 |
Evidence key: Observed
Rows are ordered by the value Muse Spark Eval Methodology published, highest first. The source does not state whether a higher value is the better result, so this page does not either: for a benchmark that measures a rate of failure, read the table from the bottom.