Most-used AI models for workflow execution
DeepSeek V4 Flash 0423 is the most-used model for workflow execution on OpenRouter, handling 13.1% of requests classified as workflow execution over the last 7 days.
This ranks models by how often they are actually chosen for workflow execution 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 | 13.1% | 10.8% | $0.140 | 1.0M |
| 2 |  Hy3 | 8.6% | 12.7% | $0.132 | 262K |
| 3 | DeepSeek V4 Pro | 7.8% | 7.0% | $0.435 | 1.0M |
| 4 |  GLM 5.2 | 5.5% | 6.8% | $0.760 | 1.0M |
| 5 |  MiMo-V2.5 | 5.2% | 7.9% | $0.140 | 1.1M |
| 6 | DeepSeek V4 Flash 0423 | 4.3% | 7.0% | $0.140 | 1.0M |
| 7 |  GPT-5.6 Luna | 3.7% | 5.8% | $0.100 | 1.1M |
| 8 |  Step 3.7 Flash | 2.6% | 3.4% | $0.200 | 262K |
| 9 | nemotron-3-ultra-550b-a55b-20260604:free | 2.4% | 4.9% | — | — |
| 10 |  Gemini 2.5 Flash Lite | 2.4% | 0.30% | $0.100 | 1.0M |
Reading the two columns together: MiMo-V2.5, 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 workflow execution jobs rather than quick one-shot calls.
Frequently asked
DeepSeek V4 Flash 0423 is the most-used model for workflow execution on OpenRouter, handling 13.1% of requests classified as workflow execution over the last 7 days. It is followed by Hy3 (8.6%) and DeepSeek V4 Pro (7.8%).
No. This ranks by how often each model is actually chosen for workflow execution 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.
Workflow Execution accounts for 7.3% of classified requests and 21.1% 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.