Anthropic
AI Safety Company, Entwickler von Claude
Claude Code
codingTerminal-basierter KI-Coding-Agent mit MCP-Integration, Computer Use und Scheduled Tasks.
WebsiteClaude.ai
productivityKI-Assistent mit Opus 4.6 (80.8% SWE-bench), 1M Token Kontext, Agent Teams.
WebsiteClaude Computer Use / Dispatch
agentsDesktop-Agent der Apps oeffnet, Browser steuert. Dispatch sendet Aufgaben per Telefon.
WebsiteAnthropic Claude 4.6
modelsOpus 4.6 (80.8% SWE-bench), Sonnet 4.6, Haiku 4.5. 1M Kontext. API Opus $5/$25.
WebsiteWhen you’re prototyping locally with AI agents like Gemini CLI, Claude Code, or your own agent, thei...
<p><a href="https://www.heise.de/news/Anthropic-KI-Sicherheitsmodell-nur-fuer-die-USA-11257464.html?wt_mc=rss.red.ho.ho....
<p><a href="https://www.heise.de/blog/Ein-Claude-Code-Plug-in-in-einem-Nachmittag-Was-ich-dabei-gelernt-habe-11229448.ht...
<p>I just got an email from Anthropic telling me they are deprecating their good model, which actually works well, claud...
<h2>What's changed</h2> <ul> <li>Updated to parity with Claude Code v2.1.109</li> </ul> <h2>Update</h2> <div class="high...
arXiv:2604.12076v1 Announce Type: new Abstract: The Identifiable Victim Effect (IVE) $-$ the tendency to allocate great...
arXiv:2604.11978v1 Announce Type: new Abstract: Large language model (LLM) agents perform strongly on short- and mid-ho...
arXiv:2604.12102v1 Announce Type: new Abstract: We introduce compute-grounded reasoning (CGR), a design paradigm for sp...
arXiv:2604.12162v1 Announce Type: new Abstract: The rapid deployment of AI agents in commercial settings has outpaced t...
arXiv:2604.12290v1 Announce Type: new Abstract: Current LLM agent benchmarks, which predominantly focus on binary pass/...