Pricing · ArbitrageUpdated 2026-10-03 · 29 providers
DeepSeek V4.1 Flash · Price by Provider
The cheapest offer (Relace, $0.020/M input) is 96% below the most expensive (Fireworks, $0.45/M). Model details
| Provider | Precision | Input /M | Output /M | Uptime 24h |
|---|---|---|---|---|
| Relacecheapest | · | $0.020 | $0.60 | 99.7% |
| OpenInference | fp4 | $0.026 | $0.78 | 94.9% |
| Morph | fp8 | $0.027 | $0.38 | 99.0% |
| InferenceNet | · | $0.040 | $0.60 | 99.8% |
| Wafer | · | $0.050 | $0.60 | 100.0% |
| Sail Research | fp4 | $0.080 | $0.40 | 99.8% |
| Decart | fp4 | $0.090 | $0.18 | 99.9% |
| Ionstream | · | $0.10 | $1.15 | 98.0% |
| DekaLLM | · | $0.12 | $0.40 | 99.5% |
| DeepInfra | fp8 | $0.14 | $0.42 | 99.9% |
| AtlasCloud | fp8 | $0.14 | $0.56 | 98.8% |
| StreamLake | fp8 | $0.14 | $0.56 | 99.5% |
| DeepSeek | · | $0.15 | $0.60 | 100.0% |
| DigitalOcean | · | $0.18 | $0.72 | 99.8% |
| GMICloud | fp8 | $0.18 | $0.72 | 99.8% |
| CoreWeave | fp8 | $0.20 | $0.65 | 98.9% |
| Makora | fp8 | $0.20 | $0.99 | 99.9% |
| NextBit | fp8 | $0.21 | $0.84 | 100.0% |
| Phala | · | $0.21 | $0.84 | 99.7% |
| Novita | fp8 | $0.24 | $0.96 | 99.5% |
| Alibaba | · | $0.30 | $1.20 | 99.7% |
| Baidu | fp8 | $0.30 | $1.20 | 99.7% |
| BaseTen | fp8 | $0.30 | $1.20 | 99.8% |
| Fireworks | · | $0.30 | $1.20 | 99.4% |
| Modal | · | $0.30 | $1.20 | 99.7% |
| Parasail | fp8 | $0.30 | $1.20 | 99.7% |
| SiliconFlow | fp8 | $0.30 | $1.20 | 99.9% |
| Together | · | $0.30 | $1.20 | 100.0% |
| Venice | fp8 | $0.30 | $1.20 | 99.8% |
| Fireworks | · | $0.45 | $1.80 | 100.0% |
OpenRouter provider list prices, USD per million tokens, collected daily · all models with price spreads