Most-used AI models for research & reports
DeepSeek V4 Flash 0423 is the most-used model for research & reports on OpenRouter, handling 12.4% of requests classified as research & reports over the last 7 days.
This ranks models by how often they are actually chosen for research & reports 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 |  DeepSeek V4 Flash 0423 | 12.4% | 11.4% | $0.140 | 1.0M |
| 2 |  Gemini 2.5 Flash | 5.9% | 1.0% | $0.300 | 1.0M |
| 3 | DeepSeek V4 Pro | 5.4% | 7.4% | $0.435 | 1.0M |
| 4 |  Hy3 | 5.0% | 10.7% | $0.132 | 262K |
| 5 |  GLM 5.2 | 4.6% | 5.0% | $0.760 | 1.0M |
| 6 | Gemini 3.5 Flash Lite | 4.1% | 0.90% | $0.300 | 1.0M |
| 7 | Gemini 2.5 Flash Lite | 3.2% | 0.40% | $0.100 | 1.0M |
| 8 | DeepSeek V4 Flash 0423 | 3.1% | 6.6% | $0.140 | 1.0M |
| 9 |  GPT-5.6 Luna | 3.1% | 5.9% | $0.100 | 1.1M |
| 10 |  Step 3.7 Flash | 3.0% | 4.4% | $0.200 | 262K |
Reading the two columns together: Hy3, DeepSeek V4 Flash 0423, GPT-5.6 Luna take a noticeably larger share of tokens than of requests — meaning they are being used for the longer, heavier research & reports jobs rather than quick one-shot calls.
Frequently asked
DeepSeek V4 Flash 0423 is the most-used model for research & reports on OpenRouter, handling 12.4% of requests classified as research & reports over the last 7 days. It is followed by Gemini 2.5 Flash (5.9%) and DeepSeek V4 Pro (5.4%).
No. This ranks by how often each model is actually chosen for research & reports 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.
Research & Reports accounts for 0.60% of classified requests and 1.1% of classified tokens on OpenRouter over the trailing 7 days.
Other tasks
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.