Pricing · ArbitrageUpdated 2026-10-05 · 26 providers
DeepSeek V4 Flash 0731 · Price by Provider
The cheapest offer (Relace, $0.015/M input) is 97% below the most expensive (Phala, $0.44/M). Model details
| Provider | Precision | Input /M | Output /M | Uptime 24h |
|---|---|---|---|---|
| Relacecheapest | fp4 | $0.015 | $1.28 | 99.8% |
| Sail Research | fp4 | $0.019 | $0.30 | 99.4% |
| Sail Research | fp4 | $0.019 | $0.42 | 99.0% |
| OpenInference | fp4 | $0.020 | $1.66 | 99.7% |
| Reka | · | $0.021 | $0.53 | 100.0% |
| StreamLake | fp8 | $0.044 | $0.13 | 99.8% |
| Inceptron | fp4 | $0.050 | $0.65 | 100.0% |
| DeepInfra | fp8 | $0.060 | $0.18 | 99.9% |
| DigitalOcean | · | $0.12 | $0.24 | 99.9% |
| BaseTen | fp8 | $0.13 | $0.26 | 100.0% |
| CoreWeave | fp8 | $0.13 | $0.28 | 100.0% |
| Cohere | · | $0.14 | $0.28 | 99.9% |
| Parasail | fp8 | $0.14 | $0.28 | 99.7% |
| Together | · | $0.14 | $0.28 | 99.7% |
| Wafer | · | $0.14 | $0.84 | 100.0% |
| Morph | bf16 | $0.14 | $0.40 | 93.8% |
| Venice | · | $0.17 | $0.35 | 99.6% |
| Mancer 2 | fp8 | $0.20 | $0.60 | 99.5% |
| SiliconFlow | fp8 | $0.22 | $0.66 | 99.5% |
| GMICloud | fp8 | $0.29 | $0.86 | 100.0% |
| Alibaba | · | $0.35 | $1.06 | 99.0% |
| NextBit | fp8 | $0.35 | $1.06 | 100.0% |
| Novita | fp8 | $0.41 | $1.23 | 100.0% |
| AtlasCloud | fp4 | $0.44 | $1.32 | 99.9% |
| Baidu | fp8 | $0.44 | $1.32 | 100.0% |
| Cloudflare | · | $0.44 | $1.32 | 100.0% |
| Phala | · | $0.44 | $1.32 | 99.9% |
OpenRouter provider list prices, USD per million tokens, collected daily · all models with price spreads