Benchmark Registry

DeepSearchQA

DeepSearchQA results reported for 5 models across 1 version. Updated 2026-10-02.

Report an issue

Versions

1 version · Metric: F1 score

VersionReleasedModelsResults
DeepSearchQA 900-prompt setLatestDecember 11, 202559

Recently reported results: DeepSearchQA 900-prompt set

Latest version: DeepSearchQA 900-prompt set

ModelScoreSource
Muse Spark 1.3 (max)90.3%dev.meta.aiF1 scoreReported Model pageMetaSelf-reportedRecord
Muse Spark 1.2 (xhigh)85.9%dev.meta.aiF1 scoreReported Model pageMetaSelf-reportedRecord
Muse Glimmer 30B (high)74.6%dev.meta.aiF1 scoreReported Model pageMetaSelf-reportedRecord
Claude Opus 5 (medium)92.8%www-cdn.anthropic.comF1 scoreReported System cardAnthropicSelf-reportedRecord
Claude Opus 5 (max)95.0%www-cdn.anthropic.comF1 scoreReported System cardAnthropicSelf-reportedRecord