GitHub - Wan-Video/Wan2.1: Wan: Open and Advanced Large-Scale Video Generative Models
💜 Wan   ｜ Â Â ðŸ–¥ï¸ GitHub   |   🤗 Hugging Face  |   🤖 ModelScope  |   📑 Technical Report   |   📑 Blog   |   💬 WeChat Group  |   📖 Discord Wan: Open and Advanced Large-Scale Video Generative Models In this repository, we present Wan2.1, a comprehensive and open suite of video foundation models that pushes the boundaries of video generation. Wan2.1 offers these key features: 👠SOTA P
Open Source Data Labeling | Label Studio
GenAI LLM Fine-Tuning Label data for supervised fine-tuning or refine models using RLHF LLM Evaluations Response moderation, grading, and side-by-side comparison RAG Evaluation Use Ragas scores and human feedback Quick Start Computer Vision Image Classification Put images into categories Object Detection Detect objects on image, boxes, polygons, circular, and keypoints supported Semantic Segmentat
Qwen2.5 Omni: See, Hear, Talk, Write, Do It All!
March 27, 2025 · 3 min · 496 words · Qwen Team | Translations:简体ä¸æ–‡ QWEN CHAT HUGGING FACE MODELSCOPE DASHSCOPE GITHUB PAPER DEMO DISCORD We release Qwen2.5-Omni, the new flagship end-to-end multimodal model in the Qwen series. Designed for comprehensive multimodal perception, it seamlessly processes diverse inputs including text, images, audio, and video, while delivering real-time streaming respo
複数㮠MCP サーãƒãƒ¼ã‚’使ã£ãŸ OpenAI Agents SDK コード例
ã™ã§ã«ã”å˜çŸ¥ã®æ–¹ã‚‚多ã„ã‹ã¨æ€ã„ã¾ã™ãŒã€OpenAI ã® Agents SDK ã‹ã‚‰ç›´æŽ¥ Model Context Protocol (MCP) サーãƒãƒ¼ã‚’ツールã¨ã—ã¦åˆ©ç”¨ã§ãるよã†ã«ãªã‚Šã¾ã—ãŸï¼å·¥å¤«æ¬¡ç¬¬ã§ã„ã‚ã„ã‚ã¨å¯èƒ½æ€§ãŒåºƒãŒã‚Šãã†ã§ã™ã。 ã“ã®æ©Ÿèƒ½ã®åŸºæœ¬çš„ãªä½¿ã„方を知るã«ã¯ Agents SDK ã®ãƒ‰ã‚ュメントページ（英語） ã¨ãƒªãƒã‚¸ãƒˆãƒªå†…ã® examples/mcp/ é…下ãŒå‚考ã«ãªã‚Šã¾ã™ãŒã€å®Ÿéš›ã« MCP サーãƒãƒ¼ã®é€£æºã‚’利用ã™ã‚‹å ´åˆã€è¤‡æ•°ã® MCP サーãƒãƒ¼ã‚’åŒæ™‚ã«åˆ©ç”¨ã™ã‚‹ã“ã¨ãŒå¤šã„ã®ã§ã¯ãªã„ã‹ã¨æ€ã„ã¾ã™ã€‚ã“ã®è¨˜äº‹ã§ã¯ã€ãã®æ–¹æ³•ã«ã¤ã„ã¦è£œè¶³è§£èª¬ã—ãŸã„ã¨æ€ã„ã¾ã™ã€‚ 紹介ã™ã‚‹ã‚³ãƒ¼ãƒ‰ä¾‹ã§ã¯ MCP プãƒã‚¸ã‚§ã‚¯ãƒˆãŒæä¾›ã—ã¦ã„る以下ã®ã‚µãƒ¼ãƒãƒ¼ã‚’利用ã—ã¾ã™ï¼š Filesystem MCP サーãƒãƒ¼ï¼šãƒãƒ¼ã‚«ãƒ«ã®ãƒ•ã‚¡ã‚¤ãƒ«ã‚·ã‚¹ãƒ†ãƒ ã¨ã®ã‚„ã‚Šã¨ã‚ŠãŒå¯èƒ½ Slack MCP サーãƒãƒ¼ï¼šSlack ワーク
日本語ã‹ã‚‰ã‚¤ãƒ©ã‚¹ãƒˆç”Ÿæˆãƒ—ãƒãƒ³ãƒ—トã«ç¿»è¨³ã™ã‚‹ LLM を作ã£ãŸ
TL;DR 日本語ã‹ã‚‰ danbooru ã‚¿ã‚°ã«ç¿»è¨³ã™ã‚‹ LLM を作ã£ãŸ LLM を使ã£ã¦ãƒ‡ãƒ¼ã‚¿ã‚»ãƒƒãƒˆã‚’åˆæˆã—㟠æˆæžœç‰© Danbooru tags Translator ã§ç•¥ã—㦠Danbot ã§ã™ (Danbooru + Bot ã«ã‚‚ãªã‚Šã¾ã™)。å‰å›žã¯ Dart ã§ã—ãŸãŒæµçŸ³ã«åå‰è¢«ã‚ŠãŒå¤šãã¦æ¤œç´¢ã—ãšã‚‰ãã†ãªã®ã§å¤‰ãˆã¾ã—ãŸã€‚ モデル: 作æˆã—ãŸãƒ‡ãƒ¼ã‚¿ã‚»ãƒƒãƒˆ: ComfyUI カスタムノード: çµæžœã ã‘見ãŸã„方㯠#ç”»åƒç”Ÿæˆçµæžœ ã¸ã€‚ å‰æ ã“ã®è¨˜äº‹ã¯ã€å‰å›žã® プãƒãƒ³ãƒ—トã¯è€ƒãˆãŸããªã„ã‘ã©ç”»åƒç”ŸæˆãŒã—ãŸã„ï¼ ã¨ãªã£ã¦ã„ã¾ã™ã€‚ 割ã¨å‰å›žã®æ‰‹æ³•ã«ã¤ã„ã¦è§¦ã‚Œã‚‹ã“ã¨ãŒã‚ã‚‹ã®ã§ã€å‰å›žã®è¨˜äº‹ã‚’èªã¾ãªã„ã¨ç†è§£ã§ããªã„箇所ãŒã‚ã‚‹ã‹ã‚‚ã—ã‚Œã¾ã›ã‚“。 ã¯ã˜ã‚ã« å‰ç½®ãã¯å‰å›žã®è¨˜äº‹ã¨åŒã˜ãªã®ã§ã–ã£ãり説明ã—ã¾ã™ã€‚ ç”»åƒç”Ÿæˆãƒ¢ãƒ‡ãƒ«ã®ãƒ—ãƒãƒ³ãƒ—トã€è€ƒãˆã‚‹ã®é›£ã—ã„ ã˜ã‚ƒã‚プãƒãƒ³ãƒ—トを LLM ã§ç”Ÿæˆã—よㆠã¨ã„ã†æ„Ÿã˜
GitHub Actions ã‚’é™çš„検査ã™ã‚‹ãƒ„ールã®ç´¹ä»‹ (actionlint/ghalint/zizmor)
先日〠tj-actions/changed-files ã‚„ reviewdog/action-* ãªã©ã®ã‚¢ã‚¯ã‚·ãƒ§ãƒ³ã® Git ã‚¿ã‚°ãŒæ›¸ãæ›ãˆã‚‰ã‚Œã‚‹ã¨ã„ã†å‡ºæ¥äº‹ãŒã‚ã‚Šã¾ã—ãŸã€‚ ã“ã‚Œã«ã‚ˆã‚Šã€ã“れらã®ã‚¢ã‚¯ã‚·ãƒ§ãƒ³ã‚’ Git ã‚¿ã‚°ã§å‚ç…§ã—ã¦ã„ã‚‹ GitHub Actions Workflow 内ã§æ‚ªæ„ã®ã‚るコードãŒå®Ÿè¡Œã•ã‚Œã¦ã—ã¾ã†ã¨ã„ã†äº‹æ…‹ãŒç™ºç”Ÿã—ã¾ã—ãŸã€‚ ã“ã®ã‚ˆã†ãªäº‹æ…‹ã‚’防ããŸã‚ã«ã¯ã€ã‚¢ã‚¯ã‚·ãƒ§ãƒ³ã®å‚ç…§ã«ã¯ Git ã‚¿ã‚°ã§ã¯ãªãコミットãƒãƒƒã‚·ãƒ¥ã‚’使用ã™ã‚‹ãªã©ã®å¯¾ç–ãŒå¿…è¦ã§ã™ã€‚ # ⌠Git ã‚¿ã‚°ã¯æ›¸ãæ›ãˆã‚‰ã‚Œã‚‹å¯èƒ½æ€§ãŒã‚ã‚‹ - uses: actions/checkout@v4 - uses: actions/[email protected] # ╠コミットãƒãƒƒã‚·ãƒ¥ã‚’指定ã—ã¦ãŠã‘ã° Git ã‚¿ã‚°ãŒæ›¸ãæ›ãˆã‚‰ã‚Œã¦ã‚‚影響をå—ã‘ãªã„ - uses: actions/checkout@11bd71901
Ruffã«æ載ã•ã‚Œã‚‹æ–°ã—ã„åž‹ãƒã‚§ãƒƒã‚«ãƒ¼Red-knotã«ã¤ã„ã¦
Python開発者å‘ã‘ツールRuff/uvã®é–‹ç™ºãƒãƒ¼ãƒ （Astral inc.）ãŒPythonå‘ã‘ã®åž‹ãƒã‚§ãƒƒã‚«ãƒ¼ã‚’æ–°è¦é–‹ç™ºã—ã¦ã„ã‚‹ã®ã§ç´¹ä»‹ã„ãŸã—ã¾ã™ã€‚ã“ã®åž‹ãƒã‚§ãƒƒã‚«ãƒ¼ã¯ã‚³ãƒ¼ãƒ‰ãƒãƒ¼ãƒ "Red-knot"ã¨å‘¼ã°ã‚Œã¦ãŠã‚Šã€ã¾ã リリースå‰ã§ã™ãŒå°†æ¥çš„ã«ã¯Ruffã«çµ„ã¿è¾¼ã¾ã‚Œã‚‹äºˆå®šã§ã‚るよã†ã§ã™ã€‚以下ã§ã¯ã“ã®ãƒ—ãƒã‚¸ã‚§ã‚¯ãƒˆã®æ¦‚略を説明ã—ãŸã„ã¨æ€ã„ã¾ã™ã€‚ 特徴 æ—¢å˜ã®åž‹ãƒã‚§ãƒƒã‚«ãƒ¼(e.g. mypy, pyright)ã®ä¸æº€ç‚¹ã§ã‚ã‚‹é…ã•ã‚’改善ã™ã¹ãã€å¾¹åº•çš„ã«ãƒ‘フォーマンスã«æ°—ã‚’é…ã£ãŸè¨è¨ˆã¨ãªã£ã¦ã„ã¾ã™ã€‚pyrightã‚‚ãã“ã¾ã§é…ãã¯ãªã„ã®ã§ã™ãŒã€ã•ã‚‰ã«é«˜é€Ÿã§ã‚ã‚‹ã“ã¨ã‚’目指ã™ã‚ˆã†ã§ã™ã€‚ Red-knotã¯Ruffã¨åŒæ§˜Rustを用ã„ã¦å®Ÿè£…ã•ã‚Œã¦ã„ã¾ã™ã€‚RuffãŒä½¿ã£ã¦ã„る諸々ã®ãƒ‡ãƒ¼ã‚¿æ§‹é€ を共有ã—ã¦ã„ã‚‹ãŸã‚ã€ãƒ•ã‚©ãƒ¼ãƒžãƒƒãƒˆã€lintã€é™çš„解æžãŒä¸€æ°—通貫ã§è¡Œãˆã‚‹ã‚ˆã†ã«ãªã‚Šã¾ã™ã€‚JavaScriptツールãƒã‚§ã‚¤ãƒ³
Google シート㸠Gemini ã®ã‚«ã‚¹ã‚¿ãƒ AI 関数を組ã¿è¾¼ã‚€
Google シート㸠Gemini ã®ã‚«ã‚¹ã‚¿ãƒ AI 関数を組ã¿è¾¼ã‚€ Google シートã§ã‚‚ã€ç”Ÿæˆ AI Gemini ã®çµ„ã¿è¾¼ã¿ ãŒæ€¥ãƒ”ッãƒã§é€²ã‚“ã§ã„ã¾ã™ã€‚ 先日多数㮠Blog ã®è©•ä¾¡ã‚’è¡Œã†æ©Ÿä¼šãŒã‚ã£ãŸéš›ã«ã¯ã€Google シート㨠Gemini ã®çµ„ã¿åˆã‚ã›ã§ã€ã‹ãªã‚ŠåŠ¹çŽ‡çš„ã«è©•ä¾¡ã‚’è¡Œã†ã“ã¨ãŒå‡ºæ¥ã¾ã—ãŸã€‚ 当時㯠GAS を利用ã—ã¦ã€åŠ›æŠ€ã§å¯¾å¿œã—ãŸã®ã§ã™ãŒã€é–¢æ•°ã¨ã—㦠Gemini を呼ã³å‡ºã›ã‚‹ã‚ˆã†ã«ã™ã‚Œã°ã€ã‚‚ã£ã¨åŠ¹çŽ‡çš„ãªåˆ©ç”¨ãŒå‡ºæ¥ã‚‹ã‚ˆã†ã«ãªã‚‹ã‹ã¨æ€ã„ã¾ã—ãŸã€‚ 残念ãªãŒã‚‰ Google シートã§ã¯ã¾ã ãã®ã‚ˆã†ãªé–¢æ•°ã®æä¾›ã¯ã•ã‚Œã¦ã„ãªã„ã®ã§ã™ãŒã€å…¬å¼ãªæä¾›ã¾ã§å¾…ã¦ãªã„ã¨ã„ã†æ–¹ã®ãŸã‚ã«ã€æœ¬è¨˜äº‹ã§ã¯ Gemini ã®ã‚«ã‚¹ã‚¿ãƒ AI 関数ã®çµ„ã¿è¾¼ã¿ã‚’試ã—ã¦ã¿ã¾ã—ãŸã€‚ ãªãœã‚¹ãƒ—レッドシートã§Geminiを使ã†ã®ã‹ï¼Ÿ スプレッドシートã¨AIã®çµ„ã¿åˆã‚ã›ã¯ã€ãƒ‡ãƒ¼ã‚¿åˆ†æžã€ãƒ¬ãƒãƒ¼ãƒˆä½œæˆã€ã‚³ãƒ³ãƒ†ãƒ³ãƒ„生
èªã¿å–り専用 DB ã‚’ Aurora ã‹ã‚‰ SQLite ã«ç§»è¡Œã—ã¦ã‚³ã‚¹ãƒˆã‚’ 1/8 ã«å‰Šæ¸›ã—ãŸè©± - エムスリーテックブãƒã‚°
ã“ã‚“ã«ã¡ã¯ã€‚クラウド型電åカルテã§ã‚るエムスリーデジカルã®ã‚½ãƒ•ãƒˆã‚¦ã‚§ã‚¢ã‚¨ãƒ³ã‚¸ãƒ‹ã‚¢å…¼ Team SRE ã‚’ã—ã¦ã„る井上 渉（@wtr_in）ã§ã™ã€‚ã‚ãƒ£ãƒ™ãƒ„ç›¸å ´ãŒè½ã¡ç€ã„ã¦ãã¦ä¸€å®‰å¿ƒã—ã¦ã„ã¾ã™ã€‚ 今回ã¯ã€ãƒ‡ã‚¸ã‚«ãƒ«ã‚’構æˆã™ã‚‹ã‚µãƒ¼ãƒ“ス㮠DB（基本的ã«èªã¿å–ã‚Šã®ã¿ï¼‰ã‚’ Aurora MySQL ã‹ã‚‰ Fargate 上㮠SQLite ã«ç§»è¡Œã—ã€æ€§èƒ½ã‚‚å‘上ã—ã¤ã¤å½“該サービス全体ã®ã‚¤ãƒ³ãƒ•ãƒ©ã‚³ã‚¹ãƒˆã‚’ç´„ 1/8 ã¾ã§å¤§å¹…ã«å‰Šæ¸›ã§ããŸè©±ã‚’ã”紹介ã—ã¾ã™ã€‚ 移行å‰ãƒ»ç§»è¡Œå¾Œã®æ§‹æˆã¨ãã®åŠ¹æžœ ç§»è¡Œå‰ (Aurora MySQL) 移行後 (SQLite) 移行ã§å¾—られãŸåŠ¹æžœ 移行を検討ã—ãŸèƒŒæ™¯ ãªãœ SQLite ã‚’é¸ã‚“ã ã‹ SQLite ã®ãƒ‡ãƒ¼ã‚¿ã‚’ã©ã“ã«ç½®ã㋠インメモリ SQLite ã¸ã®ãƒ‡ãƒ¼ã‚¿ãƒãƒ¼ãƒ‰ ã¾ã¨ã‚ We are hiring! 移行å‰ãƒ»ç§»è¡Œå¾Œã®æ§‹æˆã¨ãã®åŠ¹æžœ 今回 SQLite を採用ã—ãŸã‚µãƒ¼ãƒ“ス
リリースã€éšœå®³æƒ…å ±ãªã©ã®ã‚µãƒ¼ãƒ“スã®ãŠçŸ¥ã‚‰ã›
最新ã®äººæ°—エントリーã®é…ä¿¡
j次ã®ãƒ–ックマーク
kå‰ã®ãƒ–ックマーク
lã‚ã¨ã§èªã‚€
eコメント一覧を開ã
oページを開ã
GitHub - gusye1234/nano-graphrag: A simple, easy-to-hack GraphRAG implementation
https://openai.com/index/introducing-4o-image-generation/
kyutai/moshika-vis-pytorch-bf16 · Hugging Face
MoshiVis — Teaching Moshi to Converse about Images
DeNA Data/ML Engineering Night - 資料一覧 - connpass
https://x.com/norahsakal/status/1898183864570593663
https://ppc.land/content/files/2025/01/Newwhitepaper_Agents2.pdf
Inside Google’s Co-Scientist, Copyright Office Weighs Generated Works, and more...
https://platform.openai.com/docs/guides/agents
GitHub - mannaandpoem/OpenManus: No fortress, purely open ground. OpenManus is Coming.
SEAL LLM Leaderboards: Expert-Driven Private Evaluations
{{#tags}}- {{label}}
{{/tags}}