Wild to see an Anthropic model so far to the left on the cost/intelligence charts. This is looking like a great release!
Claude SNS 最新フィード — 48 件の最新発信
/claude-accounts/ に掲載中の Claude 関連発信者から、 各アカウントの最新投稿 1 件 を時系列で並べたページです。 毎朝 04:35 JST に更新します。
📅 最終スキャン: 2026-10-08 04:47 JST
🔐 認証モード: authenticated
📊 対象: 49 アカウント
✅ 取得対応: X
⚠️ 取得未対応(次回 PR): YouTube / note / Blog / Substack / Bluesky / GitHub / LinkedIn
Microsoft just announced that Copilot will hand routine coding from Anthropic's Claude Haiku 4.5 to Microsoft's MAI-Code-1.1-Flash (a 137B total and 6.8B active param model) running on Nvidia-powered PCs
A model router will keep th routine work on the machine and this local
For reference, Haiku 4.5 came out on October 15th, 2025 so these two columns are less than a year apart.
and oh yeah Haiku 5.5 is also much faster and 75% cheaper...
Haiku 5.5 is a good Haiku.
With 100k token context, it's also 10x cheaper than Haiku 4.5
Your backtest looks brilliant. Did your agent peek at tomorrow's news?
Backtest with Ryan Ahern of @ExaAILabs on Oct 12 at AI Engineer New York.
Search yesterday's web with Exa Snapshot.
Exa is a Platinum sponsor.
Workshops are a conference add-on.
https://ai.engineer/nyc
Small but mighty!
We've heard how much people love Haiku and it's back in 5.5, cheaper, faster than ever before.
In fact it's 10x cheaper than Haiku 4.5 under 100k tokens.
Try it with computer use, workflows and in the API.
Claude Haiku 5.5 is priced at $0.10/M input and $0.50/M output tokens, or $0.50/M and $2.50/M above 100k input tokens.
Claude Sonnet 5.5 cache reads are also now $0.10/M tokens, down from $0.20/M.
Compare Haiku 5.5 on CursorBench:
For prompts under 100K tokens, it's $0.10 / $0.50 per million tokens with cache reads at $0.01.
Above 100K tokens, $0.50 / $2.50 per million, with cache reads at $0.05.
Much of the world is based on the assumption that the future is like the past and an original meaning of “singularity” was a point where that relationship breaks. I don’t know if there will be one big Singularity, but a million little singularities across fields seems inevitable.
Another model joins the 5.5 family with the small but mighty Haiku 5.5! It’s built for speed and efficiency, and costs 75% less than Haiku 4.5. Opus does the heavy thinking, Haiku does the high-volume work, and together you get more done for less. Happy shipping!
Claude Haiku 5.5, our latest model:
Haiku 5.5 is a significant step up over Haiku 4.5 across coding, computer use, and knowledge work.
Not enough people give their AI systems goals.
I don’t mean task-specific goals (ex: make this newsletter draft under 500 words).
I mean big North Star goals (ex: I want to own a dog in the next year).
If AI is going to be a proactive support system, make it support you in
OpenAI’s Dots launched last week and they've taken over @every.
A few things that stand out:
They protect your attention. Over the weekend, my Dot posted my model-testing results to Slack and brought back replies. I could keep working without opening Slack and disappearing down
■ 補足
要約はユーザーの過去のやり取りを踏まえてパーソナライズされるのが特徴。会議メモ→計画書更新→フォローアップ作成まで一気通貫で回せる。利用はデスクトップアプリのプラグインディレクトリで「Meetings」を検索して有効化する流れ。WindowsやモバイルについてはOpenAIから言及なし。
■
OpenAI’s math release is, in the words of Opus, “the single most consequential mathematical release ever” *
LLMs have how made substantial progress on 4 of 7 Millenium Prize problems: Navier-Stokes (claimed), Riemann, Hodge and Birch-Swinnerton-Dyer. Poincaré was solved in 2003.
The full workflow and prompt that I posted on The Rundown's Community Workflow Hub:
All 4.5 years of the war in Ukraine through satellite fire detections. You can see the border line up clearly in waves. Staggering to see it from this point of view.
99 personal assistants on the wall..
99 personal assistants…
one launched today, everyone’s excited,
98 personal assistants on the wall…
AIスロップ、こういうだけ流れてくればいいのに
It is increasingly clear that humanity will solve the Riemann hypothesis before we pass the Physical Turing Test - you come back to a clean home after a wild Sunday party, and you can't tell whether a human or a robot did the job.
We are not Moravec's paradox-pilled enough.
しゃべった
はじまりました > ProtoPediaの時間 Vol.303 #NT東京 振り返り w/ @AirpocketRobot @karaage0703
わたし学生の頃は死ぬほどコミュ障でしたが、特に社会人になってから意識的に訓練して大分マシになった気がする。対人スキルは後天的に鍛えられるから、AIに仕事奪われる前にがんばろう
なんか大袈裟に言うと世界の終わりに立ち会ってるみたいな感じがあって、AIで人類滅亡ってこういう感じなんだと分からされた
弊社、自分も試したけど、普通にAstraが回路基板作り始めている。仕事として。
※Full Astraとは言ってない
ローカルLLMのAgentic Search入門｜npaka @npaka123
> 日本について「世界最悪のコンピュータセキュリティ体制を持つ国のひとつ」としたうえで、「日本の企業や政府機関に対する攻撃は今後、増加する可能性がある。これは脅しではなく、私たちの専門家としての見解」と強調しました。
【独自】“日本は世界最悪のコンピュータセキュリティ体制”
いや、マジでClaude Code × Unreal Engineやばいな。
https://x.com/rewind02/status/2107086047985258639/video/1…
最初に動画見た瞬間、「ゲーム作りもついにここまで来たか」ってなったわ。
海外の開発者が、ブループリントを1つも手で組まずに遊べるゲームを作ってる。
進め方がマジで異次元
・Unreal Engine
サイトもすごいし、めちゃくちゃ有益な知見だ
what is your default/workhorse coding agent today, Oct 2026?
I released an LLM plugin for sending prompts to OpenAI's new Jev-clone decision model that's based on GPT-6 Luna
We haven't stopped growing since we launched our new ChatGPT plugin last Tuesday.
I've never seen anything like this, we added 500K in ARR this week only!
This is also just a straight up great writing prompt - what would it mean to steal something in a post-singularity future? Who is the team? What are the security systems? How do the police work? great stuff! Session was chatham house so I can't name who said it.
the high frequency personal agent wars of 2027, fought via low latency chips and rack placement next to the servers hosting Resy and SF tennis court systems
Cool eval. Simply ask an LLM “Land or Water?” and give it a latitude and longitude coordinate as text. Ask 16,200 times, plot as image. The models know. From compressing the internet.
Tantalum being discovered in 1802 and turning out to have an isotope with a half life so long that we might never see it decay is a great example of nominative determinism.
Sonnet 5.5 is here! In Claude Code, users get ~30% more tasks done compared to Sonnet 5. It's smarter, so it needs fewer tokens for the same work.
Here it rakes a yard of leaves using tool calls. 24s faster than Sonnet 5, with 6K fewer tokens. It leaves her extra time to read :)
Anthropic just posted a write up on how they made the Claude frontend faster which legit coincidentally uses the same agentic optimization techniques as described in the blog post. https://claude.dev/blog/how-we-made-claude-ai-faster/…
Today we announced the Claude-led discovery of a molecular machine that we suspect could represent a new gene editing mechanism. Its precise function, biotechnological utility (if any), or level of significance is not yet clear, but at minimum it is work I would have been proud
very happy that claude is fun to talk to again :)
also very excited that genuinely good aesthetics are now starting to make their way into models that aren’t the fable/astra class.
universal basic aesthetics - the agi world doesn’t have to be slop!
需要あるのかわからないですがトーク後は質疑とサイン会もあります！
着席はPeatixで予約、立ち見は予約なしでOKみたいです。
ㅤ
会場で会えるの楽しみにしてます！！
ㅤ
申込（無料）：
https://1786012250.peatix.com/view
詳細：
https://store.kinokuniya.co.jp/event/1786012250/…
If you liked this, you might like my weekly newsletter. 65,000+ operators read it to sharpen their ability to build, execute, and grow.
It's free, and I publish new essays on Wednesdays at 8am ET:
I’m so proud of the amazing team we’ve assembled at Anthropic and the research we’ve done to date.
このあと17:00-から DATAFLUCTさん主催「Data Cross Conference」で久米村さん、のぐりゅーさんのトークセッションにて、AINOW編集長のおざけん @ozaken_AI がモデレーターとして出演します
よかったらご覧ください〜
https://dcc2022.datafluct.com
編集ポリシー / 取得仕様
- 掲載対象は /claude-accounts/ に登録された Claude 関連発信者のみ
- 各アカウントの ピン留め除く最新 1 件 を抽出(エンゲージメント数値は X 表示の集計)
- 更新頻度: 毎朝 04:35 JST
- X 以外(YouTube / note / Blog / Substack / Bluesky / GitHub / LinkedIn)は次の改修で取得対応予定
編集: Clauder Navi 編集部