Liquid Inference is a third-party gateway that routes to many upstream models through one API. Gateway quotes are often among the lowest on the site, but availability, rate limits and routing can vary between models, so a low headline price is worth verifying against your own workload.
It lists 8 models on this page. 7 of them have a public price. Coverage spans text and chat (8). The lowest-priced model hosted here is DeepSeek V4.1 Flash 1017.
| Cheapest here? | |||||
|---|---|---|---|---|---|
| DeepSeek V4.1 Flash 1017deepseek-v4.1-flash-10171.8× pricier | $0.059 | $0.21 | $0.0026 | 1,048,576 | 1.8× pricier |
| DeepSeek V4.1 Flashdeepseek-v4.1-flash2.2× pricier | $0.087 | $0.21 | $0.007 | 1,048,576 | 2.2× pricier |
| GLM-5.3-Flashglm-5.3-flash3.3× pricier | $0.11 | $0.32 | $0.028 | 1,048,576 | 3.3× pricier |
| Qwen3.8 27Bqwen3.8-27b1.7× pricier | $0.08 | $0.53 | $0.053 | 262,144 | 1.7× pricier |
| GLM-5.3glm-5.34.1× pricier | $0.84 | $3.10 | $0.20 | 1,048,576 | 4.1× pricier |
| Kimi K3.1 Prokimi-k3.1-pro1.3× pricier | $2.18 | $5.46 | $1.23 | 1,048,576 | 1.3× pricier |
| Kimi K3kimi-k31.9× pricier | $1.26 | $10.20 | $0.32 | 1,048,576 | 1.9× pricier |
| Liquid Autoliquid.autoOnly channel | — | — | — | 10,000,000 | Only channel |