--- title: Mistral Medium 3.5: 128B open weights model with remote... description: Mistral has released Medium 3.5, a 128B dense model with a 256k context window scoring 77.6% on SWE-Bench Verified. Weights are available on Hugging Face under... canonical: https://daily.dev/posts/mistral-medium-3-5-128b-open-weights-model-with-remote-coding-agents-ng5cn0vux twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Mistral Medium 3.5: 128B open weights model with remote coding agents | daily.dev og:description: Mistral has released Medium 3.5, a 128B dense model with a 256k context window scoring 77.6% on SWE-Bench Verified. Weights are available on Hugging Face under... og:url: https://daily.dev/posts/mistral-medium-3-5-128b-open-weights-model-with-remote-coding-agents-ng5cn0vux og:image: https://api.daily.dev/og/posts/Ng5cN0vUX.png og:image:alt: Mistral Medium 3.5: 128B open weights model with remote coding agents og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Mistral Medium 3.5: 128B open weights model with remote coding agents **[Collections](https://daily.dev/sources/collections)** · 2 min read · 2 upvotes · 0 comments ## Summary Mistral has released Medium 3.5, a 128B dense model with a 256k context window scoring 77.6% on SWE-Bench Verified. Weights are available on Hugging Face under a modified MIT license, enabling self-hosting. The Vibe coding tool now runs agents remotely in the cloud, supporting parallel sessions, GitHub/Linear/Jira/Sentry/Slack integrations, and async PR generation — putting Mistral in direct competition with OpenAI Codex, Cursor, and Claude Code. Le Chat gains a Work Mode preview for multi-step agentic workflows with user-approval guardrails. API pricing is $1.50/$7.50 per million tokens; the open weights option is the main differentiator over cheaper alternatives like Gemini Flash. ## Content Mistral has released Mistral Medium 3.5, a 128-billion parameter dense model with a 256k token context window. It scores 77.6% on SWE-Bench Verified and is now the default model in both Vibe CLI and Le Chat. ## Model details Mistral Medium 3.5 has configurable reasoning effort and is available via API at $1.50 per million input tokens and $7.50 per million output tokens. The weights are on Hugging Face under a modified MIT license, which makes self-hosting viable for teams that want to avoid API costs. ## Remote agents in Mistral Vibe The bigger shift here is what Mistral is doing with Vibe. Previously a local coding tool, Vibe now runs agents remotely in the cloud. That means you can kick off multiple sessions in parallel, install dependencies, generate PRs, and walk away while the work happens asynchronously. Agents connect to GitHub, Linear, Jira, Sentry, and Slack, and can be launched from either the CLI or directly from Le Chat. This puts Mistral in direct competition with OpenAI Codex, Cursor, and Claude Code. The open weights angle is the main differentiator — if you'd rather run the model yourself than pay per token indefinitely, that option exists. ## Work Mode in Le Chat Le Chat also gets a new Work Mode (currently in preview), which enables multi-step agentic workflows across connected tools. The intended use cases include research tasks, inbox triage, and cross-tool coordination. Sensitive operations require user approval before execution, which is a reasonable guardrail for anything touching production systems. ## Community reaction Reception has been mostly positive, though some users have flagged pricing concerns when comparing against alternatives like Gemini Flash. That's a fair point — the API rates are competitive but not the cheapest option available, so the value proposition leans more on the open weights and the integrated agent tooling than on raw cost. ## Similar posts on daily.dev - [Mistral, Europe’s answer to OpenAI and Anthropic, pushes its coding agents to the cloud](https://daily.dev/posts/mistral-europe-s-answer-to-openai-and-anthropic-pushes-its-coding-agents-to-the-cloud-ni63qasg2) · The New Stack · 1 upvotes · 0 comments --- Tags: [#llm](https://daily.dev/tags/llm), [#mistral-ai](https://daily.dev/tags/mistral-ai) [View this post on daily.dev](https://daily.dev/posts/mistral-medium-3-5-128b-open-weights-model-with-remote-coding-agents-ng5cn0vux) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Mistral Medium 3.5: 128B open weights model with remote coding agents","url":"https://daily.dev/posts/mistral-medium-3-5-128b-open-weights-model-with-remote-coding-agents-ng5cn0vux","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/mistral-medium-3-5-128b-open-weights-model-with-remote-coding-agents-ng5cn0vux"},"datePublished":"2026-04-29T16:52:48.603Z","dateModified":"2026-05-05T10:24:45.070Z","description":"Mistral has released Medium 3.5, a 128B dense model with a 256k context window scoring 77.6% on SWE-Bench Verified. Weights are available on Hugging Face under...","image":"https://pbs.twimg.com/media/HHFVj38aYAAV7vY.jpg","thumbnailUrl":"https://pbs.twimg.com/media/HHFVj38aYAAV7vY.jpg","isAccessibleForFree":true,"articleSection":"Collections","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Collections","logo":"https://media.daily.dev/image/upload/s--fk_6ycEi--/f_auto,q_auto/v1780996001/logos/collections?_a=BAMAMiWQ0","url":"https://daily.dev/sources/collections"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/mistral-medium-3-5-128b-open-weights-model-with-remote-coding-agents-ng5cn0vux","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":2},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"llm,mistral-ai","timeRequired":"PT2M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Collections","item":"https://daily.dev/sources/collections"},{"@type":"ListItem","position":3,"name":"Mistral Medium 3.5: 128B open weights model with remote coding agents"}]} ```