Official site · 🇨🇳 China
9 endpoints, cheapest first. This is the view the model-by-model tables cannot give you: what one provider actually offers, and at what precision.
| Model | Quant | $/M in | $/M out | Context | Uptime 30m |
|---|---|---|---|---|---|
| DeepSeek V4 Pro | fp8 | $0.13 | $0.40 | 1,048,576 | 100.0% |
| GLM-5.2 | fp8 | $0.14 | $0.44 | 1,048,576 | 100.0% |
| GLM-5.3 | fp8 | $0.16 | $0.49 | 1,048,576 | 99.5% |
| DeepSeek V4.1 Flash | fp8 | $0.14 | $0.56 | 1,048,576 | 99.0% |
| DeepSeek V3.2 | fp8 | $0.28 | $0.42 | 131,072 | 100.0% |
| GLM-5.2 | fp4 | $0.23 | $0.81 | 1,048,576 | 99.9% |
| DeepSeek V4 Flash | fp8 | $0.44 | $1.32 | 1,048,576 | 100.0% |
| Kimi K2.6 | fp4 | $0.43 | $1.83 | 262,144 | 99.9% |
| GLM-5.1 | fp8 | $0.96 | $3.03 | 202,752 | 99.3% |