Deepseek V4 Pro
DeepSeek 对话模型，适合代码、推理与长上下文任务。
All systems normal1 providers available
Context128K
Max output32K
$0.66/ 1M tokensInput price · by access tier0.88折$0.058/ 1M tokens2折$0.13/ 1M tokens4.5折$0.3/ 1M tokens7折$0.46/ 1M tokens$1.98/ 1M tokensOutput price · by access tier0.88折$0.17/ 1M tokens2折$0.4/ 1M tokens4.5折$0.89/ 1M tokens7折$1.39/ 1M tokensCache read$0.022/1M tokens
Cache write$0.825/1M tokens
Capabilities
Input modality
InputText
OutputText
Core capabilities
Streaming
Structured outputTool capabilities
Function callingAccess protocols
—
Pricing details
| Tier | Price (input–output) | Cache read | Cache write |
|---|---|---|---|
| ≦200K | $0.66-$1.98/1M tokens | $0.022/1M tokens | $0.825/1M tokens |
Service performance
Success rate0%Last 24 hours
First token latency2263msLast 24 hours
Quickstart
from openai import OpenAI
client = OpenAI(
base_url="https://api.apione.ai/v1",
api_key="<YOUR_API_KEY>", # 在控制台创建 API Key 后替换
)
response = client.chat.completions.create(
model="deepseek-v4-pro",
messages=[{"role": "user", "content": "用一句话介绍你自己"}],
)
print(response.choices[0].message.content)