Devstral 2

适合工程 Copilot、重构与仓库级工作流的 Mistral 编码模型。

73%+ SWE-bench公开模型详情Dense Transformer

参数规模

123B

上下文

128K

最大输出

32K

许可证

Modified MIT

TTFT

160ms

吞吐

120 tok/s

为什么选它

工程场景导向
具备较强 SWE-bench 表现

价格与差价

层级标准缓存SiliconFlow节省

在线调用$0.06 / $0.25$0.021暂无暂无

批处理$0.03 / $0.13$0.021暂无暂无

快速开始

OpenAI 兼容接口，切换 base_url 即可。

在 Playground 中试用查看价格页

Python

from openai import OpenAI

client = OpenAI(
    base_url="https://api.luminapath.tech/v1",
    api_key="BATCHIN_API_KEY"
)

resp = client.chat.completions.create(
    model="devstral-2",
    messages=[{"role": "user", "content": "Summarize why this model is a fit for my workload."}]
)

print(resp.choices[0].message.content)

JavaScript

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.luminapath.tech/v1",
  apiKey: process.env.BATCHIN_API_KEY,
});

const resp = await client.chat.completions.create({
  model: "devstral-2",
  messages: [{ role: "user", content: "Summarize why this model is a fit for my workload." }],
});

console.log(resp.choices[0]?.message?.content);

cURL

curl https://api.luminapath.tech/v1/chat/completions \
  -H "Authorization: Bearer $BATCHIN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "devstral-2",
    "messages": [{"role":"user","content":"Summarize why this model is a fit for my workload."}]
  }'

规格与行为

架构

Dense Transformer

厂商家族

BatchIn Exclusive

上下文

128K

最大输出

32K

适用场景

coding

exclusive