{"id":7326,"date":"2025-08-25T06:42:00","date_gmt":"2025-08-24T21:42:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=7326"},"modified":"2025-08-23T21:03:57","modified_gmt":"2025-08-23T12:03:57","slug":"command-a-reasoning-deepseek-v3-1-gemma-3-270m-nemotron-nano-2-dream-7b","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=7326","title":{"rendered":"Command A Reasoning, DeepSeek V3.1, Gemma 3 270M, Nemotron Nano 2, Dream 7B"},"content":{"rendered":"\n<p>LLM\/LRM\u95a2\u9023\u306e\u8a71\u984c\u306f\u672c\u5f53\u306b\u591a\u3044\u3002\u5148\u9031\u306f<a href=\"https:\/\/docs.cohere.com\/docs\/command-a-reasoning\">Cohere&#8217;s Command A Reasoning Model | Cohere<\/a>\uff08\u30e2\u30c7\u30eb\u306f<a href=\"https:\/\/docs.cohere.com\/docs\/command-a-reasoning\">Cohere&#8217;s Command A Reasoning Model | Cohere<\/a>\u3001CC-BY-NC\uff09\u306e\u516c\u958b\u3001DeepSeek V3.1\u306e\u516c\u958b\uff08<a href=\"https:\/\/api-docs.deepseek.com\/news\/news250821\">DeepSeek-V3.1 Release | DeepSeek API Docs<\/a>\u3001\u30e2\u30c7\u30eb\u306f<a href=\"https:\/\/huggingface.co\/deepseek-ai\/DeepSeek-V3.1\">deepseek-ai\/DeepSeek-V3.1 \u00b7 Hugging Face<\/a>\uff09\u304c\u5927\u304d\u306a\u30cb\u30e5\u30fc\u30b9\u3060\u3063\u305f\u3002\u30d5\u30ed\u30f3\u30c6\u30a3\u30a2\u307e\u305f\u306f\u305d\u308c\u306b\u8fd1\u3044\u30e2\u30c7\u30eb\u304c\u516c\u958b\u3055\u308c\u308b\u610f\u7fa9\u306f\u5927\u304d\u3044\u3002\u307e\u305f\u3001Intern-S1\u304b\u3089\u306f\u30c6\u30af\u30cb\u30ab\u30eb\u30ec\u30dd\u30fc\u30c8\u304c\u516c\u958b\u3055\u308c\u3066\u3044\u308b\u3002<\/p>\n\n\n\n<p>a<\/p>\n\n\n\n<p>\u5c0f\u578b\u30e2\u30c7\u30eb\u95a2\u9023\u3067\u3082Gemma 3 270M\uff08<a href=\"https:\/\/developers.googleblog.com\/en\/introducing-gemma-3-270m\/\">Introducing Gemma 3 270M: The compact model for hyper-efficient AI &#8211; Google Developers Blog<\/a>\u3001\u30e2\u30c7\u30eb\u306f<a href=\"https:\/\/huggingface.co\/google\/gemma-3-270m\">google\/gemma-3-270m \u00b7 Hugging Face<\/a>\uff09\u306f\u8d85\u5c0f\u578b\u3067\u3042\u308b\u3053\u3068\u304c\u8208\u5473\u6df1\u3044\u3002\u6027\u80fd\u7684\u306b\u306f\u7591\u554f\u304c\u3042\u308b\u3068\u306f\u3044\u3048\u7279\u5316\u7528\u9014\u306bPost training\u3059\u308b\u306a\u3069\u4f7f\u3048\u308b\u5834\u9762\u306f\u3042\u308a\u305d\u3046\u3002NVIDIA \u306eMemtron Nano2\u3082\u6ce8\u76ee\u3067\u3042\u308b\uff08Nano\u3068\u3044\u3046\u540d\u524d\u30679B\uff09\u3002<\/p>\n\n\n\n<p>Huawei\u304b\u3089\u306fDiffusion\u7cfb\u306eDream 7B\u306e\u8ad6\u6587\u304c\u51fa\u3066\u3044\u305f\u3002LLaDA\u3092\u8d85\u3048\u3001\u540c\u898f\u6a21\u306eAutoregressive\u306a\u30e2\u30c7\u30eb\u306b\u8ca0\u3051\u3066\u3044\u306a\u3055\u305d\u3046\u3068\u9ad8\u3044\u6027\u80fd\u3002<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Intern-S1: A Scientific Multimodal Foundation Model\u00a0<\/strong>[185.4]<br>Intern-S1\u306f\u3001\u4e00\u822c\u7684\u306a\u7406\u89e3\u3068\u63a8\u8ad6\u6a5f\u80fd\u3092\u5099\u3048\u305f\u5c02\u9580\u7684\u306a\u30b8\u30a7\u30cd\u30e9\u30ea\u30b9\u30c8\u3067\u3042\u308b\u3002 Intern-S1\u306f\u30aa\u30d5\u30e9\u30a4\u30f3\u304a\u3088\u3073\u30aa\u30f3\u30e9\u30a4\u30f3\u5f37\u5316\u5b66\u7fd2(RL)\u3092InternBootCamp\u3067\u5b9f\u65bd\u3059\u308b\u3002 Intern-S1\u306f\u3001\u30aa\u30fc\u30d7\u30f3\u30bd\u30fc\u30b9\u30e2\u30c7\u30eb\u9593\u306e\u4e00\u822c\u7684\u306a\u63a8\u8ad6\u30bf\u30b9\u30af\u306b\u304a\u3051\u308b\u7af6\u5408\u6027\u80fd\u3092\u793a\u3059\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2508.15763v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2508.15763v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Thu, 21 Aug 2025 17:58:00 GMT)<\/li>\n\n\n\n<li><a href=\"https:\/\/devneko.jp\/wordpress\/?p=7166\">Qwen3-Coder, Intern-S1, Step-Audio2, TeleChat2 \u2013 arXiv\u6700\u65b0\u8ad6\u6587\u306e\u7d39\u4ecb<\/a>\u3067\u53d6\u308a\u4e0a\u3052\u305f\u30e2\u30c7\u30eb\u306e\u30c6\u30af\u30cb\u30ab\u30eb\u30ec\u30dd\u30fc\u30c8<\/li>\n<\/ul>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>NVIDIA Nemotron Nano 2: An Accurate and Efficient Hybrid Mamba-Transformer Reasoning Model\u00a0<\/strong>[176.4]<br>Nemotron-Nano-9B-v2\u306f\u3001\u63a8\u8ad6\u51e6\u7406\u306e\u30b9\u30eb\u30fc\u30d7\u30c3\u30c8\u3092\u5411\u4e0a\u3055\u305b\u308b\u305f\u3081\u306b\u8a2d\u8a08\u3055\u308c\u305f\u30cf\u30a4\u30d6\u30ea\u30c3\u30c9\u306eMamba-Transformer\u8a00\u8a9e\u30e2\u30c7\u30eb\u3067\u3042\u308b\u3002 Nemotron-Nano-9B-v2\u306fNemotron-H\u30a2\u30fc\u30ad\u30c6\u30af\u30c1\u30e3\u3092\u30d9\u30fc\u30b9\u306b\u3057\u3066\u304a\u308a\u3001\u5171\u901a\u306eTransformer\u30a2\u30fc\u30ad\u30c6\u30af\u30c1\u30e3\u306e\u81ea\u5df1\u4fdd\u6301\u5c64\u306e\u5927\u90e8\u5206\u3092Mamba-2\u5c64\u306b\u7f6e\u304d\u63db\u3048\u3066\u3044\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2508.14444v2\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2508.14444v2\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Thu, 21 Aug 2025 04:18:04 GMT)<\/li>\n\n\n\n<li><a href=\"https:\/\/huggingface.co\/nvidia\/NVIDIA-Nemotron-Nano-9B-v2\">nvidia\/NVIDIA-Nemotron-Nano-9B-v2 \u00b7 Hugging Face<\/a><\/li>\n<\/ul>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Dream 7B: Diffusion Large Language Models\u00a0<\/strong>[85.3]<br>\u3053\u308c\u307e\u3067\u3067\u6700\u3082\u5f37\u529b\u306a\u30aa\u30fc\u30d7\u30f3\u62e1\u6563\u5927\u8a00\u8a9e\u30e2\u30c7\u30eb\u3067\u3042\u308bDream 7B\u3092\u7d39\u4ecb\u3057\u307e\u3059\u3002 \u6211\u3005\u306e\u30e2\u30c7\u30eb\u306f\u3001\u4e00\u822c\u7684\u306a\u3001\u6570\u5b66\u7684\u3001\u30b3\u30fc\u30c7\u30a3\u30f3\u30b0\u30bf\u30b9\u30af\u306b\u304a\u3044\u3066\u3001\u65e2\u5b58\u306e\u62e1\u6563\u8a00\u8a9e\u30e2\u30c7\u30eb\u3088\u308a\u3082\u4e00\u8cab\u3057\u3066\u512a\u308c\u3066\u3044\u307e\u3059\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2508.15487v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2508.15487v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Thu, 21 Aug 2025 12:09:58 GMT)<\/li>\n\n\n\n<li>\u300cDream 7B achieves competitive performance with Qwen 2.5 on standard benchmarks (general language understanding, mathematical reasoning, and code generation) while exhibiting superior planning abilities and novel inference flexibility features that naturally emerge from the diffusion modeling paradigm.\u300d\u3068\u306e\u3053\u3068\u3002<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/DreamLM\/Dream\">GitHub &#8211; DreamLM\/Dream: Dream 7B, a large diffusion language model<\/a>\u3001\u30e2\u30c7\u30eb\u306f<a href=\"https:\/\/huggingface.co\/collections\/Dream-org\/dream-7b-68761d3d0665386f43f310c9\">Dream 7B &#8211; a Dream-org Collection<\/a><\/li>\n<\/ul>\n","protected":false},"excerpt":{"rendered":"<p>LLM\/LRM\u95a2\u9023\u306e\u8a71\u984c\u306f\u672c\u5f53\u306b\u591a\u3044\u3002\u5148\u9031\u306fCohere&#8217;s Command A Reasoning Model | Cohere\uff08\u30e2\u30c7\u30eb\u306fCohere&#8217;s Command A Reasoning &hellip; <a href=\"https:\/\/devneko.jp\/wordpress\/?p=7326\" class=\"more-link\"><span class=\"screen-reader-text\">&#8220;Command A Reasoning, DeepSeek V3.1, Gemma 3 270M, Nemotron Nano 2, Dream 7B&#8221; \u306e<\/span>\u7d9a\u304d\u3092\u8aad\u3080<\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[2],"tags":[114,223,232,293,365],"class_list":["post-7326","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-diffusion-model","tag-llm","tag-lrm","tag-oss","tag-slm"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/7326","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=7326"}],"version-history":[{"count":1,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/7326\/revisions"}],"predecessor-version":[{"id":7327,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/7326\/revisions\/7327"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=7326"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=7326"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=7326"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}