Tok/s vs cost of self-hosting deepseek-ai/DeepSeek-V4-Flash-0731 on H100:8.
Run it yourself
deepseek-v4-flash@v1 · H100:8
$ iwant launch deepseek-v4-flashResults
Dashed lines estimate the same throughput at another cloud's price for the same GPUs.
| setup | $/hour | peak tok/s | $/1M at peak | |
|---|---|---|---|---|
| deepseek-v4-flash@v1 · H100:8 GCP europe-west2-b Spot | $11.45 | 9,948 | $0.3197 | |
| deepseek-v4-flash@v1 · GCP europe-north1 spot (estimated) | $8.70 | 9,948 | $0.2431 | |
| deepseek-v4-flash@v1 · AWS us-east-1 on-demand (estimated) | $55.04 | 9,948 | $1.5369 |