--- title: Mistral releases open weights for its 1... description: Mistral has released open weights for Mistral Large 4, its 1 trillion-parameter model, on Hugging Face. Three extra weeks of reinforcement learning pushed its... canonical: https://daily.dev/posts/mistral-releases-open-weights-for-its-1-trillion-parameter-large-4-model-tkeqfdfzn twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Mistral releases open weights for its 1 trillion-parameter Large 4 model | daily.dev og:description: Mistral has released open weights for Mistral Large 4, its 1 trillion-parameter model, on Hugging Face. Three extra weeks of reinforcement learning pushed its... og:url: https://daily.dev/posts/mistral-releases-open-weights-for-its-1-trillion-parameter-large-4-model-tkeqfdfzn og:image: https://api.daily.dev/og/posts/TkeqFdfZn.png og:image:alt: Mistral releases open weights for its 1 trillion-parameter Large 4 model 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 releases open weights for its 1 trillion-parameter Large 4 model **[The Next Web](https://daily.dev/sources/tnw)** · 3 min read · 2 upvotes · 1 comments ## Summary Mistral has released open weights for Mistral Large 4, its 1 trillion-parameter model, on Hugging Face. Three extra weeks of reinforcement learning pushed its agentic coding score on Terminal-Bench 4.0 from 28% to 41%, just ahead of GLM-5.3 at 40%. It activates 52 billion parameters per task and rivals GLM-5.3 on coding while using 47% fewer tokens per task on average. On CyberGym-E2E, a test of finding and patching memory-safety bugs, it scored 79%, matching MiMo-V2.6-Pro and beating DeepSeek V4.1 Flash's 76%. It trained on 3,800 Nvidia Grace Blackwell GPUs in Europe, and CEO Arthur Mensch said Mistral has raised roughly $6 billion in three years versus hundreds of billions at other labs. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How does Mistral Large 4 compare to GLM-5.3 on agentic coding benchmarks? Mistral Large 4 scores 41% on Terminal-Bench 4.0, slightly ahead of GLM-5.3 at 40%. On Artificial Analysis' Coding Agent Index it performs comparably to GLM-5.3 but uses 47% fewer tokens per task on average, which compounds into real savings for coding agents that run for hours. _Developers choosing between open coding models can track benchmark shifts like these on daily.dev._ ### How many parameters does Mistral Large 4 activate per task and what hardware does it run on? Mistral Large 4 is a 1 trillion-parameter model that activates 52 billion parameters per task. A quantized version is also available that runs on four Nvidia B200 or B300 GPUs. The full model was trained on 3,800 Nvidia Grace Blackwell GPUs in Europe. _Teams planning open-weight model deployments can follow hardware requirement details like this on daily.dev._ ### How does Mistral Large 4 perform on cyber defense and vulnerability patching benchmarks? Mistral Large 4 scored 79% on CyberGym-E2E, a benchmark testing whether a model can find, reproduce, and patch real memory-safety bugs in open-source C and C++ projects. That matches MiMo-V2.6-Pro and beats DeepSeek V4.1 Flash, which scored 76%. Some customers already use it for automated penetration testing agents. _Security teams evaluating AI-assisted vulnerability research can keep up with results like these on daily.dev._ ## Community discussion Top comments from developers on daily.dev. **@petecapecod** · 1 upvotes > Le Chunk lives and it wants to be local now 🐈 very cool looks like they really got those numbers up with some extra RL ## Similar posts on daily.dev - [Mistral Releases Leanstral 1.5, an Open Model That Solved 587 of 672 Putnam Math Problems](https://daily.dev/posts/mistral-releases-leanstral-1-5-an-open-model-that-solved-587-of-672-putnam-math-problems-vmefgvpn5) · DevOps.com · 27 upvotes · 7 comments --- Tags: [#ai](https://daily.dev/tags/ai), [#open-source](https://daily.dev/tags/open-source), [#ai-coding](https://daily.dev/tags/ai-coding), [#nvidia](https://daily.dev/tags/nvidia), [#mistral-ai](https://daily.dev/tags/mistral-ai) [View this post on daily.dev](https://daily.dev/posts/mistral-releases-open-weights-for-its-1-trillion-parameter-large-4-model-tkeqfdfzn) ```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 releases open weights for its 1 trillion-parameter Large 4 model","url":"https://daily.dev/posts/mistral-releases-open-weights-for-its-1-trillion-parameter-large-4-model-tkeqfdfzn","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/mistral-releases-open-weights-for-its-1-trillion-parameter-large-4-model-tkeqfdfzn"},"datePublished":"2026-10-29T13:17:11.805Z","dateModified":"2026-10-30T08:15:54.347Z","description":"Mistral has released open weights for Mistral Large 4, its 1 trillion-parameter model, on Hugging Face. Three extra weeks of reinforcement learning pushed its...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/bc529a95c694b7ddced2ae067801857d?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/bc529a95c694b7ddced2ae067801857d?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"The Next Web","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":"The Next Web","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/tnw","url":"https://daily.dev/sources/tnw"},"commentCount":1,"discussionUrl":"https://daily.dev/posts/mistral-releases-open-weights-for-its-1-trillion-parameter-large-4-model-tkeqfdfzn","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":2},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":1}],"keywords":"ai,open-source,ai-coding,nvidia,mistral-ai","timeRequired":"PT3M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"The Next Web","item":"https://daily.dev/sources/tnw"},{"@type":"ListItem","position":3,"name":"Mistral releases open weights for its 1 trillion-parameter Large 4 model"}]} {"@context":"https://schema.org","@type":"WebPage","@id":"https://daily.dev/posts/mistral-releases-open-weights-for-its-1-trillion-parameter-large-4-model-tkeqfdfzn","comment":[{"@type":"Comment","text":"Le Chunk lives and it wants to be local now 🐈 very cool looks like they really got those numbers up with some extra RL","datePublished":"2026-10-29T13:40:29.733Z","url":"https://daily.dev/posts/TkeqFdfZn#c-CetcXENW5","author":{"@type":"Person","name":"Peter Cruckshank","url":"https://daily.dev/petecapecod","image":"https://media.daily.dev/image/upload/s--ZJhQyKws--/f_auto/v1721235024/avatars/avatar_A9xh33q0QoxtkGoJRCosp"},"interactionStatistic":{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1}}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/mistral-releases-open-weights-for-its-1-trillion-parameter-large-4-model-tkeqfdfzn#faq","mainEntity":[{"@type":"Question","name":"How does Mistral Large 4 compare to GLM-5.3 on agentic coding benchmarks?","acceptedAnswer":{"@type":"Answer","text":"Mistral Large 4 scores 41% on Terminal-Bench 4.0, slightly ahead of GLM-5.3 at 40%. On Artificial Analysis' Coding Agent Index it performs comparably to GLM-5.3 but uses 47% fewer tokens per task on average, which compounds into real savings for coding agents that run for hours. Developers choosing between open coding models can track benchmark shifts like these on daily.dev."}},{"@type":"Question","name":"How many parameters does Mistral Large 4 activate per task and what hardware does it run on?","acceptedAnswer":{"@type":"Answer","text":"Mistral Large 4 is a 1 trillion-parameter model that activates 52 billion parameters per task. A quantized version is also available that runs on four Nvidia B200 or B300 GPUs. The full model was trained on 3,800 Nvidia Grace Blackwell GPUs in Europe. Teams planning open-weight model deployments can follow hardware requirement details like this on daily.dev."}},{"@type":"Question","name":"How does Mistral Large 4 perform on cyber defense and vulnerability patching benchmarks?","acceptedAnswer":{"@type":"Answer","text":"Mistral Large 4 scored 79% on CyberGym-E2E, a benchmark testing whether a model can find, reproduce, and patch real memory-safety bugs in open-source C and C++ projects. That matches MiMo-V2.6-Pro and beats DeepSeek V4.1 Flash, which scored 76%. Some customers already use it for automated penetration testing agents. Security teams evaluating AI-assisted vulnerability research can keep up with results like these on daily.dev."}}]} ```