Every host for DeepSeek V4.1 Flash 1017
| Host | Blended | Input | Output | Cached | Precision | Uptime |
|---|---|---|---|---|---|---|
| Sail Research | $0.113 | $0.101 | $0.15 | $0.0067 | fp4 | 100.0% |
| DeepInfra | $0.118 | $0.078 | $0.235 | $0.0024 | fp8 | 90.6% |
| Openinference | $0.178 | $0.07 | $0.50 | $0.008 | fp4 | 99.9% |
| Relace | $0.178 | $0.07 | $0.50 | $0.008 | — | 100.0% |
| Inferencenet | $0.185 | $0.08 | $0.50 | $0.008 | fp8 | 100.0% |
| Cloudflare | $0.225 | $0.10 | $0.60 | $0.003 | — | 99.9% |
| Decart | $0.242 | $0.138 | $0.552 | $0.0028 | fp4 | 99.6% |
| Wafer | $0.245 | $0.16 | $0.50 | $0.008 | mxfp4 | 100.0% |
| Reka | $0.262 | $0.15 | $0.60 | $0.003 | — | 98.6% |
| GMI Cloud | $0.315 | $0.18 | $0.72 | $0.0036 | fp8 | 99.9% |
| Ionstream | $0.315 | $0.22 | $0.60 | $0.019 | — | 99.4% |
| Atlas Cloud | $0.341 | $0.195 | $0.78 | $0.0039 | fp8 | 99.3% |
| Morph | $0.366 | $0.288 | $0.60 | $0.025 | fp8 | 99.7% |
| Makora | $0.413 | $0.15 | $1.20 | $0.006 | fp8 | 99.7% |
| Modal | $0.438 | $0.25 | $1.00 | $0.005 | — | 94.0% |
| Baidu | $0.493 | $0.28 | $1.13 | $0.006 | fp8 | 98.3% |
| Crusoe | $0.518 | $0.29 | $1.20 | $0.007 | fp8 | 100.0% |
| Baseten | $0.525 | $0.30 | $1.20 | $0.007 | — | 100.0% |
| DeepSeek | $0.525 | $0.30 | $1.20 | $0.006 | — | 100.0% |
| Digitalocean | $0.525 | $0.30 | $1.20 | $0.006 | — | 100.0% |
| Fireworks | $0.525 | $0.30 | $1.20 | $0.006 | — | 98.2% |
| Nebius | $0.525 | $0.30 | $1.20 | $0.006 | fp8 | 87.1% |
| Novita | $0.525 | $0.30 | $1.20 | $0.006 | fp8 | 99.7% |
| Parasail | $0.525 | $0.30 | $1.20 | $0.006 | fp8 | 100.0% |
| Phala | $0.525 | $0.30 | $1.20 | $0.006 | — | 98.4% |
| Primeintellect | $0.525 | $0.30 | $1.20 | $0.006 | — | 99.9% |
| SiliconFlow | $0.525 | $0.30 | $1.20 | $0.006 | fp8 | 99.0% |
| Together | $0.525 | $0.30 | $1.20 | $0.006 | — | 100.0% |
| Alibaba Cloud | $0.578 | $0.33 | $1.32 | $0.033 | — | 98.1% |
Cheapest price over time
- Input
- Output
What your workload costs
- Sail Research$45.30
- DeepInfra$47.04
- Openinference$71.00
- Relace$71.00
- Inferencenet$74.00
- Cloudflare$90.00
- Decart$96.60
- Wafer$98.00
- Reka$105.00
- GMI Cloud$126.00
- Ionstream$126.00
- Atlas Cloud$136.50
17 more hosts cost more.
Rent GPUs or pay per token?
- Renting the GPUs
- $16,878/mo
- Break-even volume
- 149B tokens/mo
- Needed around the clock
- 56.7K tok/s
8 × H100 cost $16,878 a month. The cheapest API (Sail Research) charges $0.113 per 1M blended tokens, so the GPUs only pay off above 149B tokens a month: the box must serve 56.7K tokens a second, every second.
GPU price: cheapest on-demand offer per GPU-hour × 730 hours. No throughput is assumed: it depends on the model size, precision, batch size and serving software, so measure yours. Power, storage, egress and engineering time are not included.
More from DeepSeek
Quality data: LMArena (CC BY 4.0), Epoch AI (CC BY 4.0). Prices: OpenRouter, LiteLLM (MIT), provider pages. List prices in US dollars before discounts and taxes; other currencies converted at the European Central Bank reference rate. The figures on these pages are free to reuse under CC BY 4.0 with a link to this page.