DeepSeek V3.1 is DeepSeek's advanced hybrid reasoning model, featuring 671 billion total parameters, 37 billion activated parameters, and a 128K-token context window. It supports both reasoning and non-reasoning modes, allowing developers to optimize for either response quality or speed depending on the task. The model delivers major improvements in reasoning, coding, tool use, and agent workflows, with performance approaching DeepSeek R1 while providing faster responses and more efficient inference. It supports structured tool calling, code agents, search agents, and complex multi-step workflows, making it an excellent choice for research, software development, enterprise automation, and general-purpose AI applications.
定价
USD · 实时汇率| 定价 | USD / M |
|---|---|
| 输入 | $0.57 |
| 输出 | $1.71 |
| 命中缓存 | $0.23 |
| 图片输入 | — |
供应商
同一模型，多渠道实时比价 · 最优值绿色高亮| 供应商 | 上下文 | 最大输出 | 输入 /M | 输出 /M | 缓存 /M | 延迟 | 吞吐 |
|---|---|---|---|---|---|---|---|
| Streamlake | 128K | 64K | $0.57 | $1.71 | $0.23 | 615ms | 73 t/s |
| baidu | 128K | 16K | $0.57 | $1.71 | — | 1106ms | 42 t/s |
运行趋势
接入示例
Model Center 会在不同服务提供商之间规范化处理请求和响应，为你统一接口。
提供兼容 OpenAI 的 Completion API：既可直接调用，也可通过 OpenAI SDK 调用，一把 key 调用全部模型。
from openai import OpenAI
client = OpenAI(
base_url="https://router-integration.test.cogfoundry.ai/api/v1",
api_key="$MODEL_CENTER_API_KEY",
)
completion = client.chat.completions.create(
model="deepseek/deepseek-v3.1",
messages=[{"role": "user", "content": "9.11 和 9.8 哪个大？"}],
stream=True,
)
for chunk in completion:
if chunk.choices and chunk.choices[0].delta.content is not None:
print(chunk.choices[0].delta.content, end="", flush=True)import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://router-integration.test.cogfoundry.ai/api/v1",
apiKey: process.env.MODEL_CENTER_API_KEY,
});
const stream = await client.chat.completions.create({
model: "deepseek/deepseek-v3.1",
messages: [{ role: "user", content: "9.11 和 9.8 哪个大？" }],
stream: true,
});
for await (const chunk of stream) {
process.stdout.write(chunk.choices[0]?.delta?.content ?? "");
}curl https://router-integration.test.cogfoundry.ai/api/v1/chat/completions \
-H "Authorization: Bearer $MODEL_CENTER_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek/deepseek-v3.1",
"messages": [{"role": "user", "content": "9.11 和 9.8 哪个大？"}],
"stream": true
}'import requests
response = requests.post(
"https://router-integration.test.cogfoundry.ai/api/v1/messages",
headers={"Authorization": "Bearer $MODEL_CENTER_API_KEY", "Content-Type": "application/json"},
json={
"model": "deepseek/deepseek-v3.1",
"max_tokens": 1024,
"messages": [{"role": "user", "content": "9.11 和 9.8 哪个大？"}],
},
)
print(response.json()["content"][0]["text"])const response = await fetch("https://router-integration.test.cogfoundry.ai/api/v1/messages", {
method: "POST",
headers: {
Authorization: "Bearer $MODEL_CENTER_API_KEY",
"Content-Type": "application/json",
},
body: JSON.stringify({
model: "deepseek/deepseek-v3.1",
max_tokens: 1024,
messages: [{ role: "user", content: "9.11 和 9.8 哪个大？" }],
}),
});
const data = await response.json();
console.log(data.content[0].text);curl https://router-integration.test.cogfoundry.ai/api/v1/messages \
-H "Authorization: Bearer $MODEL_CENTER_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek/deepseek-v3.1",
"max_tokens": 1024,
"messages": [{"role": "user", "content": "9.11 和 9.8 哪个大？"}]
}'