DeepSeek V4 Flash 0423 is the most-used model for web search on OpenRouter, handling 24.3% of requests classified as web search over the last 7 days.
This ranks models by how often they are actually chosen for web search in production traffic, not by benchmark score. Adoption reflects price, latency and availability as much as raw capability — which is exactly why it often disagrees with the benchmark leaderboards, and why it is worth reading alongside them.
| # | Model | Request share | Token share | Input /M | Context |
|---|---|---|---|---|---|
| 1 | 24.3% | 24.0% | $0.140 | 1.0M | |
| 2 | 5.0% | 6.9% | $0.435 | 1.0M | |
| 3 | 4.3% | 8.3% | $0.132 | 262K | |
| 4 | 3.7% | 0.80% | $0.300 | 1.0M | |
| 5 | 3.6% | 4.5% | $0.760 | 1.0M | |
| 6 | 3.4% | 5.9% | $0.140 | 1.0M | |
| 7 | 3.4% | 7.2% | $0.140 | 1.1M | |
| 8 | 3.3% | 0.40% | $0.150 | 128K | |
| 9 | 3.0% | 4.5% | $0.100 | 1.1M | |
| 10 | 2.7% | 0.50% | $0.500 | 1.0M |
Reading the two columns together: Hy3, DeepSeek V4 Flash 0423, MiMo-V2.5 take a noticeably larger share of tokens than of requests — meaning they are being used for the longer, heavier web search jobs rather than quick one-shot calls.
DeepSeek V4 Flash 0423 is the most-used model for web search on OpenRouter, handling 24.3% of requests classified as web search over the last 7 days. It is followed by DeepSeek V4 Pro (5.0%) and Hy3 (4.3%).
No. This ranks by how often each model is actually chosen for web search in production traffic through OpenRouter — real-world adoption, which reflects price and availability as much as capability. For capability-based rankings, see the benchmark leaderboards.
Web Search accounts for 0.70% of classified requests and 1.2% of classified tokens on OpenRouter over the trailing 7 days.
Source: OpenRouter (openrouter.ai/rankings), as of 2026-08-04. Shares are of classified, sampled traffic over a trailing 7-day window; absolute volumes are not published.