Most-used AI models for classification
DeepSeek V4 Flash 0423 is the most-used model for classification on OpenRouter, handling 11.9% of requests classified as classification over the last 7 days.
This ranks models by how often they are actually chosen for classification 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 | 11.9% | 10.9% | $0.140 | 1.0M |
| 2 |  GPT-5.6 Luna | 11.0% | 11.9% | $0.100 | 1.1M |
| 3 |  Gemini 2.5 Flash Lite | 9.3% | 4.5% | $0.100 | 1.0M |
| 4 | Gemini 2.5 Flash | 5.6% | 3.1% | $0.300 | 1.0M |
| 5 | gpt-oss-120b | 5.3% | 3.8% | $0.037 | 131K |
| 6 | GPT-4o-mini | 4.5% | 1.4% | $0.150 | 128K |
| 7 | Gemini 3.1 Flash Lite | 3.6% | 2.3% | $0.250 | 1.0M |
| 8 | Gemma 4 26B A4B | 3.2% | 2.7% | $0.070 | 262K |
| 9 | Gemini 3 Flash Preview | 3.0% | 3.0% | $0.500 | 1.0M |
| 10 | Gemma 4 31B | 2.0% | 1.5% | $0.100 | 262K |
Frequently asked
DeepSeek V4 Flash 0423 is the most-used model for classification on OpenRouter, handling 11.9% of requests classified as classification over the last 7 days. It is followed by GPT-5.6 Luna (11.0%) and Gemini 2.5 Flash Lite (9.3%).
No. This ranks by how often each model is actually chosen for classification 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.
Classification accounts for 26.8% of classified requests and 7.9% 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.