Mistral Large 4(別名「Le Chonk」)是原生多模態模型,總參數約一萬億,每次運作會啟用 490 億參數。
模型採用混合專家架構,據報在 Mistral 的歐洲數據中心由零開始訓練。
公開權重推出前,網絡安全專家及政府部門預計會測試一個安全限制較少的版本。
What is Mistral AI’s Mistral Large 4 (“Le Chonk”), how was its 1 trillion-parameter mixture-of-experts model trained and designed, what canIllustrative AI-generated artwork; not a technical diagram of Mistral Large 4.
AI 提示
Create a landscape editorial hero image for this Studio Global article: What is Mistral AI’s Mistral Large 4 (“Le Chonk”), how was its 1 trillion-parameter mixture-of-experts model trained and designed, what can. Article summary: Mistral Large 4, nicknamed “Le Chonk,” is Mistral AI’s new flagship: a roughly 1-trillion-parameter, natively multimodal model available in an API preview since October 6, 2026. Mistral plans to release its weights later. Topic tags: general, general web, news, user generated. Style: premium digital editorial illustration, source-backed research mood, clean composition, high detail, modern web publication hero. Use reference image context only for broad subject, composition, and topical grounding; do not copy the exact image. Avoid: logos, brand marks, copyrighted characters, real person likenesses, fake screenshots, UI text, readable text, watermarks, charts wi
openai.com
Mistral Large 4 是 Mistral AI 最新的旗艦模型,暱稱「Le Chonk」。它有約一萬億個參數,但每次處理要求時,啟用的約為 490 億個。Mistral 於 2026 年 10 月 6 日開放 API 公開預覽,並表示計劃在本月底推出可下載的模型權重。至於它能否全面勝過其他領先模型,目前仍未有定論,效能宣稱有待更多測試驗證。12732
一萬億參數,唔代表每次都全部啟動
Mistral 稱 Large 4 是原生多模態模型,即設計上可處理文字以外的內容;公司特別提到模型具備多模態理解能力,包括視覺任務。Mistral 亦主打它在程式編寫及代理式工作流程上的用途。110
Mistral Large 4(別名「Le Chonk」)是原生多模態模型,總參數約一萬億,每次運作會啟用 490 億參數。
模型採用混合專家架構,據報在 Mistral 的歐洲數據中心由零開始訓練。
公開權重推出前,網絡安全專家及政府部門預計會測試一個安全限制較少的版本。
What is Mistral AI’s Mistral Large 4 (“Le Chonk”), how was its 1 trillion-parameter mixture-of-experts model trained and designed, what canIllustrative AI-generated artwork; not a technical diagram of Mistral Large 4.
AI 提示
Create a landscape editorial hero image for this Studio Global article: What is Mistral AI’s Mistral Large 4 (“Le Chonk”), how was its 1 trillion-parameter mixture-of-experts model trained and designed, what can. Article summary: Mistral Large 4, nicknamed “Le Chonk,” is Mistral AI’s new flagship: a roughly 1-trillion-parameter, natively multimodal model available in an API preview since October 6, 2026. Mistral plans to release its weights later. Topic tags: general, general web, news, user generated. Style: premium digital editorial illustration, source-backed research mood, clean composition, high detail, modern web publication hero. Use reference image context only for broad subject, composition, and topical grounding; do not copy the exact image. Avoid: logos, brand marks, copyrighted characters, real person likenesses, fake screenshots, UI text, readable text, watermarks, charts wi
openai.com
Mistral Large 4 是 Mistral AI 最新的旗艦模型,暱稱「Le Chonk」。它有約一萬億個參數,但每次處理要求時,啟用的約為 490 億個。Mistral 於 2026 年 10 月 6 日開放 API 公開預覽,並表示計劃在本月底推出可下載的模型權重。至於它能否全面勝過其他領先模型,目前仍未有定論,效能宣稱有待更多測試驗證。12732
一萬億參數,唔代表每次都全部啟動
Mistral 稱 Large 4 是原生多模態模型,即設計上可處理文字以外的內容;公司特別提到模型具備多模態理解能力,包括視覺任務。Mistral 亦主打它在程式編寫及代理式工作流程上的用途。110