modelgrep

Most-used AI models for security audit

Match · Updated 2026-08-04

DeepSeek V4 Flash 0423 is the most-used model for security audit on OpenRouter, handling 11.7% of requests classified as security audit over the last 7 days.

0.30%of all requests
0.60%of all tokens
7dwindow

This ranks models by how often they are actually chosen for security audit 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.

#ModelRequest shareToken shareInput /MContext
1deepseek logoDeepSeek V4 Flash 042311.7%9.4%$0.1401.0M
2z-ai logoGLM 5.211.3%16.4%$0.7601.0M
3deepseek logoDeepSeek V4 Flash 04239.8%15.0%$0.1401.0M
4deepseek logoDeepSeek V4 Pro6.9%10.6%$0.4351.0M
5moonshotai logoKimi K34.6%7.0%$3.001.0M
6anthropic logoClaude Sonnet 53.8%3.1%$2.001M
7x-ai logoGrok 4.33.6%0.40%$1.251M
8openai logogpt-oss-120b3.4%0.80%$0.037131K
9xiaomi logoMiMo-V2.52.7%5.6%$0.1401.1M
10minimax logoMiniMax M32.5%3.0%$0.3001.0M

Reading the two columns together: DeepSeek V4 Flash 0423, DeepSeek V4 Pro, Kimi K3 take a noticeably larger share of tokens than of requests — meaning they are being used for the longer, heavier security audit jobs rather than quick one-shot calls.

Frequently asked

What is the most-used AI model for security audit?

DeepSeek V4 Flash 0423 is the most-used model for security audit on OpenRouter, handling 11.7% of requests classified as security audit over the last 7 days. It is followed by GLM 5.2 (11.3%) and DeepSeek V4 Flash 0423 (9.8%).

Is this a benchmark ranking?

No. This ranks by how often each model is actually chosen for security audit 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.

How much of all AI traffic is security audit?

Security Audit accounts for 0.30% of classified requests and 0.60% 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.