Most NotebookLM users skip the configuration step entirely — these 5 settings change everything www.makeuseof.com/notebooklm-settings-configure-notebook/ Custom instructions improve every response Tell NotebookLM exactly what you want from it Response length controls how much you get Tell NotebookLM to be brief or exhaustive Selective sources focus your output Don’t let every
Seguir leyendoCategoría: Models and enterprises
OpenAI lanza el Modo Bloqueo de ChatGPT para el plan gratuito: la protección contra prompt injection llega a todos
OpenAI lanza el Modo Bloqueo de ChatGPT para el plan gratuito: la protección contra prompt injection llega a todos wwwhatsnew.com/2026/06/06/chatgpt-lockdown-mode-gratuito-prompt-injection-2026/
Seguir leyendoDeepSeek acaba de lograr algo impensable hace unos meses: ganar terreno entre empresas estadounidenses
DeepSeek acaba de lograr algo impensable hace unos meses: ganar terreno entre empresas estadounidenses www.xataka.com/robotica-e-ia/deepseek-acaba-lograr-algo-impensable-hace-unos-meses-ganar-terreno-empresas-estadounidenses DeepSeek ha encabezado en junio la lista de proveedores de software en tendencia de Ramp El dato apunta a nuevas compras de empresas estadounidenses
Seguir leyendoClaude beat Gemini on a 150-page document test, but not for the reason you’d think
Claude beat Gemini on a 150-page document test, but not for the reason you’d think www.makeuseof.com/claude-beat-gemini-on-150-page-document-not-for-reason-you-think/ The real reason I decided to run this test A recurring bottleneck with dense documents and two competing answers How I structured the comparison without tilting the odds The 150-page document, the exact same
Seguir leyendoIntroducing MAI-Voice-2 & MAI-Image-2.5
Introducing MAI-Voice-2 | Microsoft AI microsoft.ai/news/mai-voice-2/ MAI-Image-2.5 launches at No. 2 for image editing on Arena | Microsoft AI microsoft.ai/news/introducing-mai-image-2-5/ Home | Microsoft AI Humanist Superintelligence Responsible AI to empower humanity microsoft.ai/
Seguir leyendoAskSary: Real-Time Conversational AI | GPT-5, Claude, Grok & Gemini in One Workspace
AskSary — Real-Time Conversational AI | GPT-5, Claude, Grok & Gemini in One Workspace www.asksary.com/ AskSary. Every AI model. One workspace. Commencez sur GPT-5. Continuez sur Claude. Votre contexte vous suit partout.
Seguir leyendoThere’s a NotebookLM prompt that gives you a bird’s-eye view of everything you’ve uploaded
There’s a NotebookLM prompt that gives you a bird’s-eye view of everything you’ve uploaded www.makeuseof.com/notebooklm-prompt-overview-all-sources/ List every source in this notebook. For each one, give me the title, its main topic, three key points it covers, and how it relates to the other sources.
Seguir leyendoChatGPT now remembers your preferences automatically — and it actually works
ChatGPT now remembers your preferences automatically — and it actually works www.makeuseof.com/chatgpt-now-remembers-your-preferences-automatically-and-it-actually-works/
Seguir leyendoMy local LLM felt unfinished until I put a proper interface in front of it: OpenWedUI
My local LLM felt unfinished until I put a proper interface in front of it www.makeuseof.com/local-llm-felt-unfinished-until-put-proper-interface-in-front-of-it/ Ollama is great… for developers I want something prettier Setting up Open WebUI is super easy Docker pretty much makes it ready to go You can access it from any device Yes, even when
Seguir leyendoSemanticGuard: Cut LLM API costs with self-validating semantic cache
SemanticGuard | Cut LLM API costs with self-validating semantic cache www.semanticguard.dev/?ref=toolhunt.io Cut your LLM API costs. Without breaking responses. SemanticGuard caches your LLM responses with multi-layer verification and continuous AI-judged sampling. One line of code. OpenAI, Anthropic, Google.
Seguir leyendo