Mistral Medium 3.5
Mistral Medium 3.5
Introducing Mistral Medium 3.5, remote coding agents in Vibe, plus new Work mode in Le Chat for complex tasks. 隆重推出 Mistral Medium 3.5、Vibe 中的远程编码智能体，以及 Le Chat 中用于处理复杂任务的全新“工作模式”（Work mode）。
Coding agents have mostly lived on your laptop. Today we’re moving them to the cloud, where they run on their own, in parallel, and notify you when they’re done. 编码智能体过去大多运行在你的笔记本电脑上。今天，我们将它们迁移到云端，让它们能够独立、并行地运行，并在任务完成后通知你。
You can start them from the Mistral Vibe CLI or directly in Le Chat, offloading a coding task without leaving the conversation. 你可以通过 Mistral Vibe CLI 或直接在 Le Chat 中启动它们，无需离开对话即可卸载编码任务。
Powering this is Mistral Medium 3.5 in public preview, our new default model in Mistral Vibe and Le Chat, built to run for long stretches on coding and productivity work. 为这一切提供支持的是处于公开预览阶段的 Mistral Medium 3.5，这是我们在 Mistral Vibe 和 Le Chat 中的新默认模型，专为长时间运行编码和生产力工作而构建。
The new Work mode in Le Chat (Preview) extends this with a powerful agent for complex, multi-step tasks like research, analysis, and cross-tool actions. Le Chat 中的全新“工作模式”（预览版）通过一个强大的智能体扩展了这一功能，可处理研究、分析和跨工具操作等复杂的多步骤任务。
Highlights / 亮点
Mistral Medium 3.5, a new flagship model that merges instruction-following, reasoning, and coding into a single 128B dense model. Released as open weights, under a modified MIT license. Mistral Medium 3.5 是一款全新的旗舰模型，将指令遵循、推理和编码能力融合到一个 128B 参数的稠密模型中。该模型以开放权重形式发布，采用修改后的 MIT 许可证。
Strong real-world performance at a size that runs self-hosted on as few as four GPUs. 具备强大的实际应用性能，且规模适中，最少仅需四块 GPU 即可实现自托管运行。
Mistral Vibe remote agents for async coding: sessions run in the cloud, can be spawned from the CLI or Le Chat, and a local CLI session can be teleported up to the cloud. 用于异步编码的 Mistral Vibe 远程智能体：会话在云端运行，可通过 CLI 或 Le Chat 启动，本地 CLI 会话也可以“传送”到云端。
Start Mistral Vibe coding tasks in Le Chat. Sessions run on the same remote runtime and keep going while you step away. 在 Le Chat 中启动 Mistral Vibe 编码任务。会话在相同的远程运行时上运行，即使你离开，任务也会持续进行。
Work mode in Le Chat runs on a new agent, powered by Mistral Medium 3.5, that works through multi-step tasks, calling tools in parallel until the job is done. Le Chat 中的“工作模式”由 Mistral Medium 3.5 驱动的新型智能体运行，它能处理多步骤任务，并并行调用工具，直到工作完成。
Mistral Medium 3.5
Mistral Medium 3.5 is our first flagship merged model, available in public preview. It is a dense 128B model with a 256k context window, handling instruction-following, reasoning, and coding in a single set of weights. Mistral Medium 3.5 是我们首个旗舰级融合模型，现已开放公开预览。它是一个拥有 256k 上下文窗口的 128B 稠密模型，通过单一权重集即可处理指令遵循、推理和编码任务。
It performs strongly in real-world use, with self-hosting possible on as few as four GPUs. Reasoning effort is now configurable per request, so the same model can answer a quick chat reply or work through a complex agentic run. 它在实际使用中表现强劲，最少仅需四块 GPU 即可自托管。推理工作量现在可以按请求进行配置，因此同一个模型既能回答快速的聊天回复，也能处理复杂的智能体运行任务。
We trained the vision encoder from scratch to handle variable image sizes and aspect ratios. Mistral Medium 3.5 scores 77.6% on SWE-Bench Verified, ahead of Devstral 2 and models like Qwen3.5 397B A17B. 我们从零开始训练了视觉编码器，以处理各种图像尺寸和纵横比。Mistral Medium 3.5 在 SWE-Bench Verified 上得分为 77.6%，领先于 Devstral 2 以及 Qwen3.5 397B A17B 等模型。
It also has strong agentic capabilities and scores 91.4 on τ³-Telecom. The model was built for long-horizon tasks, calling multiple tools reliably, and producing structured output that downstream code can consume. 它还具备强大的智能体能力，在 τ³-Telecom 上得分为 91.4。该模型专为长周期任务而构建，能够可靠地调用多个工具，并生成可供下游代码使用的结构化输出。
It is the model that made async cloud agents in Vibe practical to ship. Mistral Medium 3.5 becomes the default model in Le Chat. It also replaces Devstral 2 in our coding agent, Vibe CLI. 正是该模型使得 Vibe 中的异步云端智能体具备了实际交付能力。Mistral Medium 3.5 现已成为 Le Chat 的默认模型，并取代了我们编码智能体 Vibe CLI 中的 Devstral 2。
Vibe remote agents / Vibe 远程智能体
From today, coding sessions can work through long tasks while you’re away. Many can run in parallel, and you stop being the bottleneck on every step the agent takes. 从今天起，编码会话可以在你离开时处理长任务。多个任务可以并行运行，你不再是智能体每一步操作的瓶颈。
You can start the cloud agents from the Mistral Vibe CLI or from Le Chat. While they run, you can inspect what the agent is doing, with file diffs, tool calls, progress states, and questions surfaced as you go. 你可以从 Mistral Vibe CLI 或 Le Chat 启动云端智能体。在它们运行期间，你可以通过文件差异、工具调用、进度状态以及随时出现的问题来检查智能体的操作。
Ongoing local CLI sessions can be teleported up to the cloud when you want to leave them running, with session history, task state, and approvals carrying across. 当你希望让本地 CLI 会话继续运行时，可以将其“传送”到云端，会话历史记录、任务状态和审批结果都会同步迁移。
Vibe sits between the systems engineering teams already use, with humans in the loop wherever they’re needed. It plugs into GitHub for code and pull requests, Linear and Jira for issues, Sentry for incidents, and apps like Slack or Teams for reporting. Vibe 位于工程团队现有的系统之间，并在需要时保持人工参与。它与 GitHub（用于代码和拉取请求）、Linear 和 Jira（用于问题跟踪）、Sentry（用于事故处理）以及 Slack 或 Teams（用于报告）等应用集成。
Each coding session runs in an isolated sandbox, including broad edits and installs. When the work is done, the agent can open a pull request on GitHub and notify you, so you review the result instead of every keystroke that produced it. 每个编码会话都在隔离的沙箱中运行，包括广泛的编辑和安装操作。工作完成后，智能体可以在 GitHub 上发起拉取请求并通知你，这样你只需审查结果，而无需关注产生结果的每一个按键。
It fits the high-volume, well-defined work that takes a developer’s time without taking their judgment: module refactors, test generation, dependency upgrades, CI investigations, as well as bug fixes. 它适用于那些占用开发者时间但不需要其判断力的高频、定义明确的工作：模块重构、测试生成、依赖升级、CI 调查以及 Bug 修复。
We use Workflows orchestrated in Mistral Studio to bring Mistral Vibe into Le Chat. We originally built this for our own in-house coding environment, then for our enterprise customers. Today the capability opens up to everyone, who can now launch coding tasks from the web. 我们使用 Mistral Studio 编排的工作流将 Mistral Vibe 引入 Le Chat。我们最初是为内部编码环境构建此功能，随后提供给企业客户。今天，该功能向所有人开放，用户现在可以直接从网页端启动编码任务。
And without being tied to a local terminal, a developer can run several in parallel. You can start coding sessions directly in Le Chat, so a task described in chat runs on the same remote runtime as the CLI and the web, and comes back later as a finished branch or a draft PR. 开发者不再受限于本地终端，可以并行运行多个任务。你可以直接在 Le Chat 中启动编码会话，这样在聊天中描述的任务就会在与 CLI 和网页端相同的远程运行时上运行，稍后会以完成的分支或草稿 PR 的形式返回。
New Work mode in Le Chat (Preview) / Le Chat 中的全新“工作模式”（预览版）
Work mode is a powerful new agentic mode for complex tasks in Le Chat, powered by a new harness and Mistral Medium 3.5. The agent becomes the execution backend for the assistant itself, so Le Chat can read and write, use several tools at once, and work through multi-step projects until it completes what you’ve asked. “工作模式”是 Le Chat 中用于处理复杂任务的强大新智能体模式，由全新的框架和 Mistral Medium 3.5 驱动。智能体成为助手本身的执行后端，因此 Le Chat 可以进行读写、同时使用多种工具，并处理多步骤项目，直到完成你的要求。
Here’s what Work mode enables you to do today: 以下是“工作模式”目前能为你实现的功能：
Cross-tool workflows: catch up across email, messages, and calendar in a single run; prepare for a meeting with attendee context, latest news, and talking points pulled from your sources. 跨工具工作流：在一次运行中同步处理电子邮件、消息和日历；通过从你的来源获取参会者背景、最新新闻和讨论要点来准备会议。
Research and synthesis: dive into a topic across the web, internal docs, and connected tools, then produce a structured brief or report you can edit before exporting or sending. 研究与综合：深入研究网络、内部文档和已连接工具中的主题，然后生成结构化的摘要或报告，供你在导出或发送前进行编辑。
Triage your inbox and draft replies; create issues in Jira from your team and customer discussions; send a summary to your team on Slack. 整理收件箱并起草回复；根据团队和客户的讨论在 Jira 中创建问题；向 Slack 上的团队发送摘要。
Sessions persist longer than a typical chat reply, so an agent can keep going across many turns, through trial-and-error, and through to completion. 会话的持续时间比典型的聊天回复更长，因此智能体可以在多次对话中持续运行，通过反复试验直至完成任务。
In Work mode, connectors are on by default rather than chosen manually, which lets the agent reach into documents, mailboxes, calendars, and other systems for the rich context it needs to take correct action. 在“工作模式”下，连接器默认开启，无需手动选择，这使得智能体能够访问文档、邮箱、日历和其他系统，获取采取正确行动所需的丰富上下文。
Every action the agent takes is visible: you see each tool call and the thinking rational 智能体采取的每一个行动都是可见的：你可以看到每一次工具调用及其思考逻辑。