# Top models

Share of token volume across every model we serve, all teams included, over the last 30 days.

Share of all token volume through the gateway, every team included, from Aug 18, 2026 to Sep 16, 2026 (UTC, 30 days). The tail past the top nine is folded into Other, as is any model whose lab this catalog does not name.

| Rank | Model | Share |
| --- | --- | --- |
| 1 | [DeepSeek V4.1 Flash](https://inference.boundless.network/models/deepseek-v4.1-flash) | 43.6% |
| 2 | [DeepSeek V4 Flash](https://inference.boundless.network/models/dsv4) | 29.7% |
| 3 | [GLM-5.2](https://inference.boundless.network/models/glm-5.2) | 11.7% |
| 4 | [GLM-5.3 Flash](https://inference.boundless.network/models/glm-5.3-flash) | 8.8% |
| 5 | [Qwen3.6-35B-A3B](https://inference.boundless.network/models/qwen3.6) | 1.6% |
| 6 | [qwen3-coder-next](https://inference.boundless.network/models/qwen3-coder-next) | 1.2% |
| 7 | [GPT-5.6 Luna](https://inference.boundless.network/models/gpt-5.6-luna) | 1.0% |
| 8 | [Qwen3.8 Flash](https://inference.boundless.network/models/qwen3.8-flash) | 0.7% |
| 9 | [Kimi K3](https://inference.boundless.network/models/kimi-k3) | 0.7% |
| 10 | Other | 1.0% |

The same window by lab is [Top labs](https://inference.boundless.network/rankings/labs). Every model's rates and capabilities are on [Models](https://inference.boundless.network/models).