[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"tag-multimodal-model":3},{"tag":4,"articles":11,"peer_article_count":44},{"id":5,"name":6,"slug":7,"article_count":8,"description_zh":9,"description_en":10},"e8dc5ded-c6ae-4bc2-a394-20e0ae5cd8d2","multimodal model","multimodal-model",3,"多模態模型把文字、圖片、程式碼與其他感知訊號放進同一個推理框架，讓模型能看圖寫碼、理解介面、處理文件與代理任務。這類模型的重點不只在能力提升，也在部署成本、上下文長度與工具整合方式。","Multimodal models combine text with vision, code, and other signals in one system, enabling tasks like image-grounded coding, UI understanding, document analysis, and agent workflows. Their real impact depends on capability, context length, and deployment cost.",[12,21,29,36],{"id":13,"slug":14,"title":15,"summary":16,"category":17,"image_url":18,"cover_image":18,"language":19,"created_at":20},"82f156ee-9163-4fec-a11b-19a123329980","claude-mythos-preview-benchmarks-price-context-en","Claude Mythos Preview: Benchmarks, Price, Context","Anthropic’s Claude Mythos Preview is an unreleased multimodal model with a proprietary license and a place above Opus.","model-release","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1779401177636-e372.png","en","2026-05-21T22:05:50.777661+00:00",{"id":22,"slug":23,"title":24,"summary":25,"category":17,"image_url":26,"cover_image":27,"language":19,"created_at":28},"674cce69-5be8-4c32-bfbd-32ab6fd2fcd7","qwen36-27b-open-source-coding-model-en","Qwen3.6-27B opens a smaller, sharper path to coding","Qwen3.6-27B is a 27B dense multimodal model that beats Qwen3.5-397B-A17B on key coding benchmarks while staying easier to deploy.","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1777260618061-cpw4.png","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Fskill-cover-qwen36-27b-en_-1777263004.png","2026-04-27T00:12:39.968514+00:00",{"id":30,"slug":31,"title":32,"summary":33,"category":17,"image_url":34,"cover_image":34,"language":19,"created_at":35},"465ff868-0532-49a3-8e00-843bf34845c7","kimi-k25-brings-vision-code-and-swarm-agents-en","Kimi K2.5 Brings Vision, Code, and Swarm Agents","Moonshot AI's Kimi K2.5 adds native vision, 256K context, and Agent Swarm. Here's what changes for developers and teams.","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1775156819505-x8wn.png","2026-04-02T19:06:37.16594+00:00",{"id":37,"slug":38,"title":39,"summary":40,"category":41,"image_url":42,"cover_image":42,"language":19,"created_at":43},"e64aba21-254b-4f93-aa21-837484bb52ec","kimi-k25-review-stronger-still-not-legend-en","Kimi K2.5 review: stronger, still not a legend","Kimi K2.5 adds vision, coding, and multi-agent tools, but long runs, weak art direction, and paywalls keep it from elite status.","ai-agent","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Fcover-1774595754485-vd5ahn.png","2026-03-27T07:15:55.385951+00:00",0]