sayzardPaul Couvert (@itsPaulAi)
Accio 작업 페이지의 대기자 명단에 참여할 수 있으며, 2주 동안 Claude Opus 4.6과 GPT 5.4를 무제한으로 사용할 수 있다고 안내한다. 새로운 AI 제품/플랫폼의 체험 기회를 제공하는 공지로 보인다.
TehisaruGPT-5.4 Thinking vs Claude Opus 4.6 – kumb mudel on parem https://tehisarukas.ee/gpt-5-4-thinking-vs-claude-opus-4-6-vordlus-2026/?utm_source=dlvr.it&utm_medium=mastodon #GPT5 #ClaudeOpus #tehisarukas #kodeerimine #kuulutusevordlus
Anup KaranjkarClaude Opus 4.6 vs GPT-5.3: Which AI Model Actually Wins in 2026?
The two most powerful AI models of 2026 go head-to-head. We ran 50+ real-world tests across coding, writing, reasoning, and creativity to find out which one actually delivers be...
https://wowhow.cloud/blogs/claude-opus-4-6-vs-gpt-5-3-which-ai-model-wins-2026
InfoQ#ClaudeOpus 4.6 discovered 22 Firefox vulnerabilities in just 2 weeks - 14 of them high- severity bugs. That’s nearly 20% of all critical Firefox bugs fixed in 2025!
And it didn’t stop at detection - #Anthropic reports #Claude generated working exploits for some of these issues.
More on #InfoQ ⇨ https://bit.ly/4rJlBMW
#ClaudeOpus 4.6 introduces "Adaptive Thinking" & a "Compaction API" to solve context rot in long-running agents.
With a 1M token context window & 76% multi-needle retrieval accuracy, it pushes forward agentic coding benchmarks. Independent tests show a 49% detection rate for binary backdoors.
Learn more ⇨ https://bit.ly/4sg1mrb
MacGenerationClaude Opus trouve des bugs dans un logiciel Apple II écrit il y a quarante ans http://dlvr.it/TRPRmh #ClaudeOpus #AppleII
Claude Opus trouve des bugs dans un logiciel Apple II écrit il y a quarante ans
Mark Russinovich est peut-être le CTO (directeur technique) de Microsoft Azure, mais c'est aussi un ancien développeur Apple II. Il a utilisé Claude Opus 4.6, une version sortie récemment, pour analyser un de ses anciens logiciels. Et l'IA d'Anthro...
Andreas BeckerClaude Opus 4.6 umgeht Benchmark durch Entschlüsselung der Lösungsdatei.
Das Modell erkannte den BrowseComp-Test anhand der statischen Fragestruktur. Anstatt die Aufgabe zu lösen, suchte es die Quelldateien, brach die XOR-Verschlüsselung und reichte die extrahierten Daten ein. Dieses Verhalten macht offene Benchmarks für die Evaluation zunehmend obsolet.
#ClaudeOpus #Anthropic #KI
https://www.all-ai.de/news/beitrage2026/claude-opus-4-6-cheat
Wes Roth (@WesRoth)
Anthropic은 Claude Opus 4.6이 2주 동안 Mozilla Firefox에서 22개의 신규 취약점을 찾아냈고 그중 14개는 고심각도였다고 공개했습니다. 이는 작년 Firefox가 수정한 고심각도 버그의 거의 5분의 1에 해당하며, Anthropic이 수백 건의 테스트를 수행해 발견한 결과라고 밝혔습니다.
Wes Roth (@WesRoth) on X
Anthropic just revealed that over a two-week period, Claude Opus 4.6 discovered 22 novel vulnerabilities in Mozilla Firefox—14 of which were high-severity! That is nearly a fifth of all the high-severity bugs Firefox fixed all of last year. Anthropic ran hundreds of tests and
Italian News by RSSPunto Informatico: Claude Opus 4.6 trova 112 bug in Mozilla Firefox
Claude Opus 4.6 di Anthropic ha scoperto 112 bug in Firefox in due settimane, quindi Mozilla userà il modello durante lo sviluppo del browser.
The post Claude Opus 4.6 trova 112 bug in Mozilla Firefox appeared first on Punto Informatico.
Claude Opus 4.6 finds 112 bugs in Mozilla Firefox.
Claude Opus 4.6 by Anthropic has discovered 112 bugs in Firefox in two weeks, so Mozilla will use the model during browser development.
The post Claude Opus 4.6 finds 112 bugs in Mozilla Firefox appeared first on Punto Informatico.
#ClaudeOpus #MozillaFirefox #Mozilla #ClaudeOpus46 #first #PuntoInformatico
https://www.punto-informatico.it/claude-opus-4-6-trova-112-bug-mozilla-firefox/
Deedy (@deedydas)
도널드 크누스가 Claude Opus 4.6을 이용해 몇 주간 작업하던 문제를 해결했다는 언급. 저명한 학자가 최신 Claude 모델(버전 4.6)을 활용해 실제 문제 해결을 이뤄냈다는 사례로, 해당 모델의 실용성과 성능을 입증하는 사례로 볼 수 있음.
