# DeepSeek: DeepSeek V4 Pro — price and measured latency by host

Updated July 25, 2026. Served by 18 providers; 7 of them measured here.

Latency and uptime are measured by llmlatency.dev probes. Prices are published by the providers and read via OpenRouter on 2026-07-25 — not measured.

| Host | Input $/M | Output $/M | Context | Fastest p50 | From | Uptime |
| --- | --- | --- | --- | --- | --- | --- |
| deepseek | $0.435 | $0.870 | 1,048,576 | 172 ms | ap-tokyo | 100% |
| novita | $1.168 | $2.336 | 1,048,576 | 71 ms | us-central | 100% |
| deepinfra | $1.300 | $2.600 | 1,048,576 | 226 ms | us-central | 99.9% |
| siliconflow | $1.502 | $3.135 | 1,048,576 | 82 ms | us-central | 100% |
| baseten | $1.740 | $3.480 | 262,144 | 54 ms | us-central | 100% |
| together | $1.740 | $3.480 | 512,000 | 111 ms | us-central | 100% |
| fireworks | $1.740 | $3.480 | 1,048,576 | 17 ms | ap-tokyo | 100% |

Source: https://llmlatency.dev/model/deepseek-deepseek-v4-pro · CC BY 4.0