Lotu RadarAbout

Latest News Archive - Page 396

World · The Guardian Ukraine

Ukraine war briefing: ‘With precision, as planned’ – Kyiv outlines battlefield gains

Zelenskyy announces recapture of 745 sq km in Oleksandrivskyi sector; plea from stranded Ukrainian frontline troops. What we know on day 1,632 Continue reading...

Business · CNBC Technology

South Korea's Kospi swings from bear to bull-market territory in just over a month on AI trade

South Korean stocks rallied on Thursday, pushing the benchmark Kospi into a technical bull market.

World · Al Jazeera

Half a tonne of Erling Haaland branded cocaine seized in Ecuador

Erling Haaland’s name and image turned up on hundreds of cocaine bricks seized by Ecuadorian police.

Developers & Open Source · Hugging Face Blog

What We Learned by Reproducing 2,200 papers from ICML

Hugging Face Blog published: What We Learned by Reproducing 2,200 papers from ICML

World · Al Jazeera

Iran war live: US, Tehran exchange opposing claims over Hormuz control

President Trump again insists US has 'total control' of strait, while Tehran resolute that the US must meet its demands.

Notable Blogs · Simon Willison

DeepSeek V4 Pro 0813 (on OpenRouter)

DeepSeek V4 Pro 0813 (on OpenRouter) The latest DeepSeek Pro model is now available, via API only. I had to link to OpenRouter because DeepSeek don't have any obvious announcement page for their new model. I haven't been able to confirm if they plan to release the open weights, but given the weights are available for both April's deepseek-ai/DeepSeek-V4-Pro and July's deepseek-ai/DeepSeek-V4-Flash-0731 it seems likely. Update : the weights are now available on Hugging Face, 1.7T parameters, 893 GB. Interestingly I got very different looking pelicans for the three different reasoning levels of low, medium, and high. I've not noticed this kind of difference from any other model: Low: Medium: High: In terms of benchmarks... as far as I can tell those were released to the Official DeepSeek WeChat Group, then copied and pasted into a post on Reddit which was deleted by the moderators for being "low-effort", then copied into this ASCII-art table on Hacker News . Tags: ai , generative-ai , llms , pelican-riding-a-bicycle , deepseek , llm-release , ai-in-china

Products & Consumer Tech · Product Hunt

Kivicube

Create AR experiences with AI, no code required! Discussion | Link

Developers & Open Source · ollama/ollama Releases

v0.32.10

What's Changed Models that don't set a repeat_penalty now default to 1.0 (off) instead of 1.1, matching other engines and speeding up speculative decoding; set a per-model parameter if an older model repeats itself. Faster prefill on NVFP4 MLX models with a global scale, about 7–8% on Qwen3.6 and Muse Glimmer. Fixed blob verification being skipped when an OCI manifest's config and layer share a digest. New Contributors @vigneshakaviki made their first contribution in #15504 Full Changelog : v0.32.8...v0.32.10-rc1

Products & Consumer Tech · Product Hunt

Phinq

Stops AI agents before they break something Discussion | Link

Developers & Open Source · vercel/next.js Releases

v16.3.1-canary.15

Misc Changes test: fix 'ouput' typo in parseIdsFromCliOutput: #97266 Add the mtime fallback back: #97268 Allow issue participants to request reopening: #96947 [turbopack] Document clusters for chunking: #97267 Link to issues in the contribution guide and discourage trivial PRs: #97188 Stop serializing hmr version state: #97146 [turbopack] Add clusters to chunking heuristics: #95157 [test] Redirect the deployment host in-process for proxy deploy tests: #97260 [test] Unflake two cache-components-dev-streaming assertions: #97246 [test] Unflake use-cache-default-profile-expire-zero tests: #97250 [turbopack] Fix HMR for dynamic imports evaluated from layouts: #97213 [ci] Use OIDC tokens to read private preview builds: #96919 Allow running ncc before pnpm build : #97234 Invalidate deferred Webpack entry sources: #97205 Credits Huge thanks to @syedsohailhussain1 , @lukesandberg , @marcoshernanz , @sampoder , @unstubbable , @eps1lon , and @gnoff for helping!

Startups & Funding · Hacker News Best

Principia Mathematica is modern and insightful

Article URL: https://okmij.org/ftp/Computation/Impressions/PrincipiaMathematica.html Comments URL: https://news.ycombinator.com/item?id=49279928 Points: 232 # Comments: 106

World · Al Jazeera

Trump administration arrests Southern Poverty Law Center expert for fraud

Heidi Beirich, an expert on far-right groups, has been charged over alleged misuse of donor funds to pay informants.

World · Al Jazeera

Media groups sue Trump over paid access to public social media posts

Lawsuit claims Trump profits unfairly by charging up to $100,000 monthly for exclusive access to presidential posts.

Products & Consumer Tech · Product Hunt

Scholé Learn by Building

Learning should be more than "click next" Discussion | Link

Developers & Open Source · Chrome Releases

Chrome for Android Update

 Hello Everyone! We've just released Chrome 152 (152.0.7977.42) for Android to a small percentage of users. It'll become available on Google Play over the next few days. You can find more details about early Stable releases here . This release includes stability and performance improvements. You can see a full list of the changes in the Git log . If you find a new issue, please let us know by filing a bug . Krishna Govind Google Chrome

Society · The Guardian Society

Nurses fainting in ‘inhumane’ conditions due to extreme heat, union warns

RCN calls for more air conditioners as Burnham faces pressure to recognise urgency of adaptation to climate crisis Nurses are collapsing during shifts amid “inhumane and impossible” conditions caused by extreme heat, their union has warned, as Andy Burnham promised to grapple with “the bigger questions” of climate adaptation. Amid growing pressure from Labour backbenches to fully acknowledge the urgency of adaptation measures as the climate crisis bites, Burnham said he was for now focusing on the immediate crises caused by an ongoing summer of extreme heat. Continue reading...

Society · BBC Technology

Why Japanese firms are being so slow to use AI

Japanese risk aversion and conservatism blamed for slow AI take-up by the country's business sector.

Products & Consumer Tech · Ars Technica

Toddler's tragic death from brain-destroying amoeba offers lessons for doctors

Boy's unusual symptoms could help identify the next case sooner.

Business · CNBC Technology

Jim Cramer says the AI data center trade is back. These 6 stocks are leading the comeback

CNBC’s Jim Cramer said the AI data center trade is regaining market leadership after weeks of forced selling weighed on the group.

AI · TechCrunch AI

Some Claude users are mad that Anthropic’s new watermarks will catch them using it at their jobs, classes

Is Anthropic's new watermarking system a travesty? Some have taken to social media to complain that it is.

World · Al Jazeera

Trump administration plans to give ICE officers electric-shock gloves

US Homeland Security is expected to buy up to $20m in gloves amid criticism of the president's immigration crackdown.

Society · NPR Top Stories

ICE plans to outfit officers with electric shock gloves

The company that makes the gloves, Compliant Technologies, says they're a nonlethal tool to make individuals compliant during arrests.

World · Deutsche Welle

Brazil orders Discord to halt livestreaming after teen death

Brazil has ordered Discord to suspend livestreaming as authorities scrutinize how the platform protects young users. The move follows a case described a "real-life horror show" that has triggered national outrage.

Products & Consumer Tech · Ars Technica

The web’s newest weapon against AI scrapers is a font

“ShieldFont” aims to poison AI training data without making pages unreadable for people.