OpenRouter Launches Web Search Benchmarks for Models and Configurations

AravSrinivas · x · 2026-08-14

OpenRouter announced Web Search Benchmarks, providing rankings of search tools across different models and configurations to help decide how to ground agents. Benchmarks include τ²-Bench Airline, GPQA Diamond, BrowseComp, DeepSearchQA, and HLE. Each score links to configuration, costs, and telemetry. For example, in BrowseComp, Perplexity with Claude Opus 5 (high) leads with 89.0% quality; in DeepSearchQA, Perplexity with Claude Opus 5 (high) leads with 76.5%.

Original post →

More from Research

Research channel →