
Managed inference API for large AI and ML models
DeepInfra has 4 plans: paid plans from $0.09. We last read its pricing page on 3 October 2026. We have not recorded a price change.
DeepInfra plans and prices
DeepSeek-V3.2
usage based$0.26
$0.32 / 1M input tokens, $0.89 / 1M output tokens
- overage$0.38($0.38 per 1M output tokens)
- usage based$0.38($0.38 per 1M output tokens)
- Model Context — 160k
- $ per 1M input tokens — $0.26
- $ per 1M output tokens — $0.38
DeepSeek-R1-0528
usage based$0.5
$0.50 / 1M input tokens ($0.35 cached), $2.15 / 1M output tokens
- overage$2.15($2.15 per 1M output tokens)
- Context window — 160k
- Output pricing — $2.15 / 1M tokens
- Cached input pricing — $0.35 / 1M tokens
- Model family — DeepSeek
DeepSeek-V4-Pro
usage based$1.30
$1.30 per 1M input tokens
- overage$0.1($0.10 per 1M cached output tokens)
- usage based$2.60($2.60 per 1M output tokens)
- Model Context — 1024k
- $ per 1M input tokens — $1.30
- $ per 1M output tokens — $2.60
DeepSeek-V4-Flash
usage based$0.09
$0.09 per 1M input tokens
- overage$0.018($0.018 per 1M cached output tokens)
- usage based$0.18($0.18 per 1M output tokens)
- Model Context — 1024k
- $ per 1M input tokens — $0.09
- $ per 1M output tokens — $0.18
Price history
Last 3 recorded changes
| Date | Plan | Change | Old | New | Δ% |
|---|
| Aug 4, 2026 | DeepSeek-V3.2 | Reclassified | $0.32 | $0.26 | — |
| Jul 27, 2026 | DeepSeek-V4-Flash | New plan | — | $0.09 | — |
| Jul 27, 2026 | DeepSeek-V4-Pro | New plan | — | $1.30 | — |
Other AI & ML products
No confirmed alternatives yet.