DeepSeek V4 Pro: API pricing, providers and specs
DeepSeek V4 Pro is a text model by DeepSeek. 33 providers serve it, from $0.418 to $4.80 /1M; the lowest price is at OpenRouter.
How DeepSeek V4 Pro compares
Against the 737 text models on the site priced the same way.
- Cheap for its class: at $0.418 /1M, its lowest price undercuts 68% of comparable models (median $1.10).
- Its 1.05M-token context window is larger than 83% of them (median 262K), about 1570 pages of text.
- 33 providers serve it; 99% of comparable models have fewer.
- Its weights are open under the MIT license: commercial use allowed.
DeepSeek V4 Pro prices by provider
Every provider's list price, read from its own price page or API. A price marked “from” changes with context length, resolution or mode.
| Provider | Input | Output | Prompt caching | Context |
|---|---|---|---|---|
| OpenRouter | $0.209/1M | From$0.418/1M | $0.017/1M | 1.05M |
| DevPass (LLM Gateway) | $0.435/1M | From$0.87/1M | $0.004/1M | — |
 Kenari | $0.552/1M | $1.10/1M | $0.006/1M | — |
 CometAPI | $0.528/1M | From$1.58/1M | — | — |
| ZenMux | $0.66/1M | $1.98/1M | $0.022/1M | 1M |
 OrcaRouter | $0.66/1M | $1.98/1M | $0.022/1M | 1.05M |
| Vercel AI Gateway | $0.66/1M | $1.98/1M | $0.022/1M | 1M |
 CrazyRouter | $0.67/1M | $2/1M | — | — |
 Opper | $1.06/1M | $2.60/1M | $0.04/1M | — |
 Hugging Face | $1.30/1M | From$2.60/1M | — | — |
 Deep Infra | $1.30/1M | From$2.60/1M | $0.10/1M | — |
| APIMart | $1.03/1M | $3.09/1M | — | — |
 Novita AI | $1.60/1M | $3.20/1M | $0.135/1M | — |
 Jiekou.AI | $1.60/1M | $3.20/1M | $0.135/1M | — |
 EmpirioLabs AI | $1.65/1M | $3.30/1M | $1.65/1M | 1M |
 Venice AI | $1.65/1M | $3.30/1M | $0.33/1M | — |
 Alibaba | $1.65/1M | $3.30/1M | — | — |
 SiliconFlow | $1.67/1M | $3.33/1M | — | — |
 AIHubMix | $1.69/1M | $3.38/1M | $0.14/1M | 1M |
| Atlas Cloud | $1.68/1M | $3.38/1M | $0.13/1M | — |
 Cortecs | $1.71/1M | $3.42/1M | $0.151/1M | — |
| OpenCode | $1.74/1M | $3.48/1M | $0.145/1M | — |
 FastRouter | $1.74/1M | $3.48/1M | $0.145/1M | 1.05M |
 Azure | $1.74/1M | From$3.48/1M | $0.145/1M | — |
 Impossibl | $1.74/1M | $3.48/1M | $0.145/1M | — |
 Requesty | $1.75/1M | From$3.50/1M | $0.44/1M | 1.05M |
 Prime Inference | $1.91/1M | $3.83/1M | — | — |
 APIYI | $1.32/1M | From$3.96/1M | — | — |
 LaoZhang API | $1.74/1M | $3.96/1M | — | — |
| BytePlus | $1.32/1M | From$3.96/1M | $0.044/1M | — |
 Poe | $1.33/1M | From$4/1M | $0.133/1M | 1.05M |
 CrossModel | $1.35/1M | $4.05/1M | $0.045/1M | — |
 Alibaba | $2.40/1M | From$4.80/1M | — | — |
- OpenRouterFrom$0.418
- DevPass (LLM Gateway)From$0.87
- Kenari$1.10
- CometAPIFrom$1.58
- ZenMux$1.98
- OrcaRouter$1.98
- Vercel AI Gateway$1.98
- CrazyRouter$2
- Opper$2.60
- Hugging FaceFrom$2.60
- Deep InfraFrom$2.60
- APIMart$3.09
Specs
- Type
- Text
- Maker
- DeepSeek
- Inputs
- Text
- Outputs
- Text
- Context
- 1.05M≈ 1570 pages of text
- Max output
- 384K
- Capabilities
- Prompt caching, JSON, Reasoning, Tool calling
- Released
- Apr 22, 2026
- License
- MITcommercial use allowed
- Open weights
- Yes, downloadable
- Parameters
- 1.6T
- Monthly searches
- 40K
Alternatives to DeepSeek V4 Pro
Same kind of model from other makers, at a similar price.
| Model | Lowest price | Context | Arena | Providers |
|---|---|---|---|---|
| DeepSeek V4 Prothis model | $0.418/1M | 1.05M | — | 33 |
 GLM-5.3 | $1.10/1M | 1.05M | 1623 | 49 |
| GLM-5.3-Flash | $0.20/1M | 1.05M | 1608 | 45 |
| Qwen3.8 27B | $0.50/1M | 262K | 1591 | 33 |
| GPT-6 Luna | $0.40/1M | 1.05M | 1578 | 28 |
 Sonar | $1/1M | 127K | — | 9 |
| Claude Haiku 5.5 | $0.30/1M | 1M | 1585 | 27 |
 MiMo-V2.6-Pro | $0.696/1M | 1.05M | 1625 | 21 |
| Qwen3.8 Flash Next | $0.376/1M | 262K | 1628 | 5 |
More from DeepSeek
DeepSeek V4.1 FlashDeepSeek$0.03/1MDeepSeek V4 FlashDeepSeek$0.17/1MDeepSeek V4 Flash Vision ExpDeepSeek$0.294/1MDeepSeek R1DeepSeek$1.79/1MDeepSeek V3DeepSeek$0.89/1MDeepSeek V4 Flash 0731DeepSeek$0.133/1MDeepSeek R1 Distill Qwen 7BDeepSeek$0.01/1MDeepSeek R1 Distill Qwen 32BDeepSeek$0.20/1MDeepSeek V4 Pro FAQ
The lowest price for DeepSeek V4 Pro is $0.418 /1M, at OpenRouter.
OpenRouter, at $0.418 /1M. 33 providers serve DeepSeek V4 Pro; the table above lists every one with its price.
DeepSeek V4 Pro reads up to 1.05M tokens per request and writes up to 384K.
Yes. DeepSeek V4 Pro's weights can be downloaded from its maker's Hugging Face page, under the MIT license.
About 1.6T, as the safetensors metadata of its official Hugging Face repo states.
33 API providers serve DeepSeek V4 Pro. Compare their prices above, or open a provider to see everything it offers.
Can't find what you're looking for? Contact us