Official site · 🇨🇳 China
14 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 Flash | fp8 | $0.09 | $0.18 | 1,024,000 | 99.8% |
| DeepSeek V3.2 | fp8 | $0.21 | $0.32 | 128,000 | 99.4% |
| Qwen3 Coder Next | unknown | $0.18 | $0.90 | 256,000 | — |
| DeepSeek V3 | unknown | $0.26 | $1.03 | 128,000 | 99.8% |
| DeepSeek V3.1 Terminus | unknown | $0.34 | $1.03 | 128,000 | 100.0% |
| KAT-Coder-Pro V2 | unknown | $0.30 | $1.20 | 256,000 | — |
| MiniMax-M3 | fp8 | $0.30 | $1.20 | 1,000,000 | 99.5% |
| Xiaomi MiMo-V2.5-Pro | unknown | $0.52 | $1.04 | 1,000,000 | — |
| GLM 4.7 | fp8 | $0.48 | $1.76 | 200,000 | 96.4% |
| Kimi K2.6 | fp8 | $0.60 | $2.52 | 256,000 | 98.5% |
| Kimi K2.5 | fp8 | $0.54 | $2.70 | 256,000 | 99.9% |
| GLM-5.2 | fp8 | $0.75 | $2.37 | 1,024,000 | 99.7% |
| Qwen3.5 397B A17B | unknown | $0.60 | $3.60 | 256,000 | 96.2% |
| GLM-5.1 | fp8 | $0.97 | $3.04 | 200,000 | 99.8% |