--- title: Moonshot releases Kimi K3.1 Pro and says it cut its... description: Moonshot released Kimi K3.1 Pro, an upgraded open-weight version of Kimi K3, with full weights due by 4 November. On Terminal-Bench 4.0, it scored 65.7%, ahead... canonical: https://daily.dev/posts/moonshot-releases-kimi-k3-1-pro-and-says-it-cut-its-hacking-skills-dsdailll6 twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Moonshot releases Kimi K3.1 Pro and says it cut its hacking skills | daily.dev og:description: Moonshot released Kimi K3.1 Pro, an upgraded open-weight version of Kimi K3, with full weights due by 4 November. On Terminal-Bench 4.0, it scored 65.7%, ahead... og:url: https://daily.dev/posts/moonshot-releases-kimi-k3-1-pro-and-says-it-cut-its-hacking-skills-dsdailll6 og:image: https://api.daily.dev/og/posts/DsDAilLL6.png og:image:alt: Moonshot releases Kimi K3.1 Pro and says it cut its hacking skills 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. # Moonshot releases Kimi K3.1 Pro and says it cut its hacking skills **[The Next Web](https://daily.dev/sources/tnw)** · 2 min read · 1 upvotes · 0 comments ## Summary Moonshot released Kimi K3.1 Pro, an upgraded open-weight version of Kimi K3, with full weights due by 4 November. On Terminal-Bench 4.0, it scored 65.7%, ahead of GPT-6 Astra's 58.2% and just behind Claude Opus 5.5's 66.4%. On Moonshot's own stress test, false claims fell to 8.3% from 23.4% for K3. Moonshot removed cyberattack-related material from training before the open release, so it would not push the limits of freely available models. On Moonshot's in-house test of exploiting real software flaws, the released model solved 81% of tasks versus 90.5% for a version trained without those cuts and 33.3% for K3, and by Moonshot's count DeepSeek's V4.1-Flash scored 85.7%. The figures are Moonshot's own and have not been checked independently. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How does Kimi K3.1 Pro compare to Claude Opus 5.5 on coding benchmarks? Kimi K3.1 Pro scored 65.7% on Terminal-Bench 4.0, a test of AI coding agents, just behind Claude Opus 5.5's 66.4% and ahead of GPT-6 Astra's 58.2%. These figures come from Moonshot itself and have not been independently verified. Full model weights for K3.1 Pro are due by November 4. _Developers choosing a coding model can track these contested benchmark claims on daily.dev._ ### Did Moonshot remove hacking capabilities from the open-weight Kimi K3.1 Pro model? Yes, Moonshot removed cyberattack-related training data before releasing the weights to limit misuse. A version trained without those cuts solved 90.5% of in-house exploit tasks, while the released model solved 81%, still far above the original K3's 33.3%. DeepSeek's open model V4.1-Flash reportedly scored 85.7% on the same measure. _Security teams weighing open-weight model risks can follow these safety trade-offs on daily.dev._ ## Similar posts on daily.dev - [Everyone Read Kimi K3 as a Price War. It Is the Opposite.](https://daily.dev/posts/everyone-read-kimi-k3-as-a-price-war-it-is-the-opposite--xshhpdjyb) · Medium · 3 upvotes · 0 comments --- Tags: [#ai](https://daily.dev/tags/ai), [#open-source](https://daily.dev/tags/open-source), [#ai-safety](https://daily.dev/tags/ai-safety), [#kimi](https://daily.dev/tags/kimi), [#moonshot-ai](https://daily.dev/tags/moonshot-ai) [View this post on daily.dev](https://daily.dev/posts/moonshot-releases-kimi-k3-1-pro-and-says-it-cut-its-hacking-skills-dsdailll6) ```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":"Moonshot releases Kimi K3.1 Pro and says it cut its hacking skills","url":"https://daily.dev/posts/moonshot-releases-kimi-k3-1-pro-and-says-it-cut-its-hacking-skills-dsdailll6","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/moonshot-releases-kimi-k3-1-pro-and-says-it-cut-its-hacking-skills-dsdailll6"},"datePublished":"2026-10-23T18:22:14.068Z","dateModified":"2026-10-23T18:53:11.734Z","description":"Moonshot released Kimi K3.1 Pro, an upgraded open-weight version of Kimi K3, with full weights due by 4 November. On Terminal-Bench 4.0, it scored 65.7%, ahead...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/03a78cffba0b3775f309590e041d9c74?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/03a78cffba0b3775f309590e041d9c74?_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":0,"discussionUrl":"https://daily.dev/posts/moonshot-releases-kimi-k3-1-pro-and-says-it-cut-its-hacking-skills-dsdailll6","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai,open-source,ai-safety,kimi,moonshot-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":"The Next Web","item":"https://daily.dev/sources/tnw"},{"@type":"ListItem","position":3,"name":"Moonshot releases Kimi K3.1 Pro and says it cut its hacking skills"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/moonshot-releases-kimi-k3-1-pro-and-says-it-cut-its-hacking-skills-dsdailll6#faq","mainEntity":[{"@type":"Question","name":"How does Kimi K3.1 Pro compare to Claude Opus 5.5 on coding benchmarks?","acceptedAnswer":{"@type":"Answer","text":"Kimi K3.1 Pro scored 65.7% on Terminal-Bench 4.0, a test of AI coding agents, just behind Claude Opus 5.5's 66.4% and ahead of GPT-6 Astra's 58.2%. These figures come from Moonshot itself and have not been independently verified. Full model weights for K3.1 Pro are due by November 4. Developers choosing a coding model can track these contested benchmark claims on daily.dev."}},{"@type":"Question","name":"Did Moonshot remove hacking capabilities from the open-weight Kimi K3.1 Pro model?","acceptedAnswer":{"@type":"Answer","text":"Yes, Moonshot removed cyberattack-related training data before releasing the weights to limit misuse. A version trained without those cuts solved 90.5% of in-house exploit tasks, while the released model solved 81%, still far above the original K3's 33.3%. DeepSeek's open model V4.1-Flash reportedly scored 85.7% on the same measure. Security teams weighing open-weight model risks can follow these safety trade-offs on daily.dev."}}]} ```