LLMs & Generative AI - Page 44 of 73
Latest breakthroughs in large language models and generative AI shaping the future of artificial intelligence and machine learning.
1459 articles
View complete article list
LWiAI Podcast #238: GPT 5.4 Mini, OpenAI Pivot, Mamba 3, Attention Residuals
The dust has barely settled on last week’s AI news cycle, and the landscape has already shifted again. OpenAI drops GPT‑5.4 Mini and Nano, faster, yes, but the price hike stings.
Meta's structured prompting lifts LLM code review accuracy to 93%
Meta’s structured prompting method now lets large language models check code for crashes without running it. The technique achieved 93% accuracy in a patch review test, the company reported.
Claude leak of 512,000+ lines reveals Tamagotchi-style pet, always‑on agent
Half a million lines of code spilled into the open. Inside the leak: a digital pet that reacts to your every keystroke, an always-on background agent called KAIROS, and an Anthropic engineer’s quiet confession that a clever optimization might have...
Anthropic's Claude Code source leak prompts guidance for users
Losing your code is bad. Losing the instruction manual for your security system is a catastrophe. That's what just happened to Anthropic. Someone appears to have dumped a massive chunk of Claude Code's source onto the internet.
Zero‑Budget Full‑Stack App Uses Free Whisper, GLM‑4.7‑Flash and FastAPI
Tech founders love to promise a free lunch. Most serve leftovers. But an actual developer just built a complete audio-summarization app without opening their wallet. OpenAI's free Whisper model handles the transcription.
Anthropic unveils secret 'Mythos' model for AI development in internal testing
Anthropic has a ghost in the machine. While the world watched its legal battles and Pentagon standoffs, the company was quietly building something else, a model it calls "Mythos." Now in internal testing, Mythos isn’t another chatbot.
IndexCache sparse attention optimizer makes long-context AI 1.82× faster
Long-context AI models have a problem with the paperwork. The real computational work of figuring out which words matter to which other words has already been solved, more or less. Dynamic sparse attention made that core task efficient.
Apple's Siri to support third‑party AI chatbots via new 'Extensions' feature
Bloomberg’s Mark Gurman reports that Apple, a company historically obsessed with control, is now wiring Siri to outside AI chatbots through a new “Extensions” feature.
Import AI memories and chat history to Gemini for seamless continuity
Google wants you to switch to Gemini. Their latest pitch isn't about smarter answers, but a better memory. You can now paste a summary of your chats from another AI, like ChatGPT, directly into Gemini's settings. The idea is simple.
Intercom's Fin Apex 1.0 beats GPT‑5.4, Claude Sonnet 4.6 in service
Generic models are impressive parlor tricks. They can write a sonnet about a missing package but often fail to actually find it. The customer service AI race is over, and the specialists just won.
Build a Smolagent with Qwen2.5-Coder-32B-Instruct using Hugging Face token
Forget those vague AI demos. You can build a functional coding agent in fifteen minutes. Today. Called Smolagents, it writes Python, executes it, and accomplishes a specific task.
Gemini 3.1 Flash Live cuts latency, boosts pitch and pace detection
Chatting with an AI still has the whiff of a parlor trick. That's changing fast. Google just launched Gemini 3.1 Flash Live, a model built to slaughter the lag between question and answer.
Anthropic ships Claude with Mac control, joins OpenClaw, NemoClaw race
The race to give AI a pair of digital hands just got a new leader. Anthropic has shipped computer control for Claude on Mac, dropping its entry into a market that OpenClaw ignited and NemoClaw is now chasing.
Tely AI offers hands‑off recommendations in ChatGPT, Google, Claude in a week
Your business is invisible in AI chat results. That can change in a week, without you lifting a finger. Tely AI flips the script: instead of you hunting for visibility, ChatGPT, Google, and Claude start recommending your company. No writers.
Google's TurboQuant boosts AI memory 8×, slashes serving costs half
AI has been about bigger, always bigger. More parameters, more data, more compute. The strategy was brute force. Now it's about less. Google's TurboQuant algorithm shrinks a model's memory footprint by eight times.
Google's TurboQuant cuts AI key‑value cache size without quality loss
Large language models are glorified pattern matchers. They don’t think; they predict. To do that convincingly, they rely on a “digital cheat sheet”, the key-value cache, that holds semantic vectors representing everything they’ve processed.
Lyria 3 supports image‑to‑music input, shaping audio in Google AI Studio
Ask a computer for sad music, and you'll likely get a predictable minor-key dirge. Feed Google's new Lyria 3 model a picture of a rusted car in a vacant lot, though. The result might be a melody you've never heard.
Anthropic introduces safer auto mode for Claude Code, balancing handholding and autonomy
Every coding assistant sells a solo act. You give the prompt; it delivers. That velocity has a price, often paid in deleted databases or secrets emailed to the wrong inbox. Now, Anthropic has built a new speed bump directly into Claude Code.
Google adds Gemini checkout partners while OpenAI upgrades ChatGPT product views
Google is cutting deals. This week, the tech giant announced a new partnership with Gap, folding the retailer into its Gemini-powered checkout system. Ask the AI for jeans, and you can buy a pair without ever leaving the chat window.
ChatLLM launches Route LLM to auto‑select models for writing, code, research
Choosing an AI model has become a trap. You can waste an entire afternoon just evaluating options. ChatLLM's new platform, Route LLM, is a direct attempt to stop that cycle. It presents a single box.
Featured Resources & Reviews
No Code MBA Course Review
Learn to build AI-powered apps without coding. Our comprehensive review of No Code MBA's course.
AI Tools & Resources
Curated collection of AI tools, courses, and frameworks to accelerate your AI journey.
Weekly AI Digest
Get the week's most important AI news delivered to your inbox every week.