Model information
Released
2026-04-24
Context window
1M
Max output
384K
Model parameters
1.6T
Weights
Open
Modalities
Input
Text
Output
Text
Identity
- Version
- deepseek-v4-pro
- Knowledge cutoff
- 2025-05
- License
- MIT
- Aliases
- DeepSeek V4 Pro
Resources
Capabilities
Open weightReasoningStructuredTemperatureTool call
Official price
input $0.435/M · output $0.87/M
cache read $0.004/M
Providers
| Provider | Provider model ID | Protocol | Context | Max output | Input | Output | Capabilities | Notes | Pricing | Streaming |
|---|---|---|---|---|---|---|---|---|---|---|
| AnyAPI | deepseek/deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | - | Yes | ||
| CrossModel | deepseek/deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $0.47/M · output $0.94/M cache read $0.005/M · cache write $0.47/M | Yes | ||
| OpenCode Go | deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $0.435/M · output $0.87/M cache read $0.004/M | Yes | ||
| Vivgrid | deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $0.435/M · output $0.87/M cache read $0.004/M | Yes | ||
| NanoGPT | deepseek/deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $1.1/M · output $2.2/M cache read $0.11/M | Yes | ||
| FastRouter | deepseek/deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $1.74/M · output $3.48/M | Yes | ||
| CrofAI | deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $0.35/M · output $0.8/M cache read $0.003/M | Yes | ||
| LLM Gateway | deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $0.435/M · output $0.87/M cache read $0.004/M | Yes | ||
| Kenari | deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $0/M · output $0/M | Yes | ||
| OpenCode Zen | deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $1.74/M · output $3.84/M cache read $0.145/M | Yes | ||
| UnoRouter | deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $0.9/M · output $1.8/M | Yes | ||
| UnoRouter | deepseek-v4-pro:free | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $0/M · output $0/M | Yes | ||
| FrogBot | deepseek-v4-pro | openai_chat | 1M | 384K | AttachmentsTool callTemperature | - | input $1.74/M · output $3.48/M cache read $0.14/M | Yes | ||
| ZenMux | deepseek/deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $0.435/M · output $0.87/M cache read $0.004/M | Yes | ||
| EmpirioLabs AI | deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $1.65/M · output $3.3/M cache read $1.65/M | Yes | ||
| Alibaba Token Plan | deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $0/M · output $0/M cache read $0/M · cache write $0/M | Yes | ||
| Venice AI | deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $1.65/M · output $3.301/M cache read $0.33/M | Yes | ||
| DigitalOcean | deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $1.392/M · output $2.784/M cache read $0.348/M | Yes | ||
| Model Oracle AI | deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | - | Yes | ||
| Auriko | deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $0.435/M · output $0.87/M cache read $0.004/M | Yes | ||
| Azure | deepseek-v4-pro | openai_responses | 1M | 384K | ReasoningStructuredTemperature | - | input $1.74/M · output $3.48/M | Yes | ||
| Ofox | deepseek/deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $0.45/M · output $0.88/M cache read $0.004/M | Yes | ||
| OrcaRouter | deepseek/deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $0.56/M · output $1.12/M cache read $0.004/M | Yes | ||
| routing.run | deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $0.348/M · output $0.696/M | Yes | ||
| DeepSeek | deepseek-v4-pro | custom | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $0.435/M · output $0.87/M cache read $0.004/M | Yes | ||
| TensorX | deepseek/deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $1.75/M · output $3.5/M cache read $0.438/M · cache write $2.185/M | Yes | ||
| Kilo Gateway | deepseek/deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $1.6/M · output $3.2/M cache read $0.135/M | Yes | ||
| Merge Gateway | deepseek/deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callTemperature | - | input $0.435/M · output $0.87/M cache read $0.004/M | Yes | ||
| Modelis | deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $0.435/M · output $0.87/M | Yes | ||
| Charm Hyper | deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $2.4/M · output $4.8/M cache read $0.2/M | Yes | ||
| Vercel AI Gateway | deepseek/deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $0.435/M · output $0.87/M cache read $0.004/M | Yes | ||
| Alibaba (China) | deepseek-v4-pro | custom | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $0.435/M · output $0.87/M cache read $0.004/M | Yes | ||
| NovitaAI | deepseek/deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $1.6/M · output $3.2/M cache read $0.135/M | Yes | ||
| OpenRouter | deepseek/deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | Reasoning levels: xhigh, high, Default reasoning: high, Hugging Face: deepseek-ai/DeepSeek-V4-Pro | input $0.435/M · output $0.87/M cache read $0.004/M | Yes | ||
| Alibaba Token Plan (China) | deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $0/M · output $0/M cache read $0/M · cache write $0/M | Yes | ||
| Ollama Cloud | deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool call | - | - | Yes | ||
| Cortecs | deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callTemperature | - | input $1.553/M · output $3.106/M cache read $0.004/M | Yes | ||
| HPC-AI | deepseek/deepseek-v4-pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $1.74/M · output $3.48/M cache read $0.145/M | Yes | ||
| EBCloud | DeepSeek-V4-Pro | openai_chat | 1M | 384K | ReasoningTool callStructuredTemperature | - | input $0.429/M · output $0.857/M | Yes |