Most-used AI models for repo scanning
MiMo-V2.5 is the most-used model for repo scanning on OpenRouter, handling 17.1% of requests classified as repo scanning over the last 7 days.
This ranks models by how often they are actually chosen for repo scanning 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 |  MiMo-V2.5 | 17.1% | 16.2% | $0.140 | 1.1M |
| 2 |  DeepSeek V4 Flash 0423 | 11.2% | 8.2% | $0.140 | 1.0M |
| 3 | DeepSeek V4 Pro | 7.8% | 5.2% | $0.435 | 1.0M |
| 4 |  Claude Sonnet 4.6 | 7.4% | 3.0% | $3.00 | 1M |
| 5 |  GLM 5.2 | 5.9% | 5.6% | $0.760 | 1.0M |
| 6 | DeepSeek V4 Flash 0423 | 5.5% | 9.6% | $0.140 | 1.0M |
| 7 |  GPT-5.6 Luna | 3.8% | 8.3% | $0.100 | 1.1M |
| 8 | GLM 5 | 3.7% | 2.1% | $0.950 | 205K |
| 9 |  Step 3.7 Flash | 3.6% | 4.1% | $0.200 | 262K |
| 10 |  Hy3 | 2.9% | 3.4% | $0.132 | 262K |
Reading the two columns together: 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 repo scanning jobs rather than quick one-shot calls.
Frequently asked
MiMo-V2.5 is the most-used model for repo scanning on OpenRouter, handling 17.1% of requests classified as repo scanning over the last 7 days. It is followed by DeepSeek V4 Flash 0423 (11.2%) and DeepSeek V4 Pro (7.8%).
No. This ranks by how often each model is actually chosen for repo scanning 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.
Repo Scanning accounts for 0.50% of classified requests and 1.5% 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.