SmolLM2 family of models (135M to 1.7B parameters) achieves strong performance on-device, making AI accessible without cloud dependencies.
Delivery couriers will be able to earn money by completing activities like filming everyday tasks or recording themselves speaking in another language.
@AMD: AMD has once again been named to the JUST 100 list by @JustCapital and @CNBC , ranking 18th among America’s 100 Most JUST Companies in 2026. We’re especially proud to be among the top 5% of companies in supporting our workforce, reflecting our continued focus on investing in our
42 points • 16 comments on Hacker News
@ElevenLabs [3/19/2026]: @ElevenCreative For creators, marketers, and businesses, it replaces the complexity of traditional music licensing - no sync fees, no per-use negotiations, no clearance delays. Select your license tier, download, and use.
@ElevenLabs [3/19/2026]: @ElevenCreative Start creating with @ElevenCreative today. https://t.co/dljjw1uWHC https://t.co/FvVyyK0b65
@ElevenLabs [3/19/2026]: Introducing the Music Marketplace in @ElevenCreative. Creators, artists, and musicians can now publish and earn from their tracks created with our music model. https://t.co/tXmoZnsvWB
Nominate your startup, or one you know, for TechCrunch Startup Battlefield 200 before May 27. Chance to win $100,000 equity-free funding and VC access.
Kaggle (@kaggle) shared: Start your own here:
We’re opening up the Kaggle toolbox to everyone. Today, we’re launching Community Hackathons - a free, self-serve way for you to host your own AI challenges. Whether you're an educator, a meetup lead, or just have a big idea, you can now build, judge and award prizes (up to
20 points • 4 comments on Hacker News
Perplexity Enterprise Pro brings AI-powered research to teams with internal knowledge search, citations, and enterprise-grade security.
@runwayml [3/19/2026]: Register now: https://t.co/iKiqgteBDS
13 points • 4 comments on Hacker News
802 points • 496 comments on Hacker News
Just feed your images into the bot and it’ll try to mimic specific styles and aesthetics. | Image: Adobe Adobe is launching customizable AI image generators that can mimic specific artistic styles and character designs. The Firefly Custom Models are available in public beta starting today, allowing creators and brands to train a model on their own assets to ensure generated images follow a consistent aesthetic for characters, illustrations, and photography. The tool aims to streamline workf...
@databricks on anthropic & claude. Agentic analytics is changing how data teams work, and it's happening faster than expected. Ken Wong from Databricks, @rgb_prithvi from @AnthropicAI , and other experts discuss how agents built on
@databricks on ai news & insights. Databricks CEO and co-founder @alighodsi is taking the stage at #RSAC2026 to address the most critical challenge in security: the shift from manual operations to machine-scale AI automation. Hear w Delta Sharing now supports Attribute-Based Access Control. Providers can now share these tables directly without creating separate copies, and recipients can apply their own ABAC policies with the gua
@vkhosla on ai regulation & policy. Congratulations @elonmusk on the win. Hopefully you can get Trump to do some of the things he said he'd do and not do some of the things he promised to do. Make the best of it now for the country. F
@saranormous on open source ai. One of my favorite @NoPriorsPod discussions: the wonderful @karpathy on - status of autonomy, tesla v. waymo - humanoid robots - bottlenecks to AGI - our future exo-brains - small models, open sou
@percyliang on ai benchmarks. Marin 32B Base (mantis) is done training! It is the best open-source base model (beating OLMo 2 32B Base) and it’s even close to the best comparably-sized open-weight base models, Gemma 3 27B PT and
@soumithchintala on ai hardware & chips. There's three parts. 1. Fitting as large of a network and as large of a batch-size as possible onto the 10k/100k/1m H100s -- parallelizing and using memory-saving tricks. 2. Communicating state bet Meta announces 2nd-gen inference chip MTIAv2. * 708TF/s Int8 / 353TF/s BF16 * 256MB SRAM, 128GB memory * 90W TDP. 24 chips per node, 3 nodes per rack. * standard PyTorch stack (Dynamo, Inductor, Trito
@amasad on deepseek. So much cope about DeepSeek. Not only did they release a great model. they also released a breakthrough training method (R1 Zero) that’s already reproducing. I doubt they lied about training costs,
2 sources discuss ai coding tools, including @amasad, @cursor_ai. Many are asking if they can move their Next.js projects from Vercel to Replit? Yes! - Go to Import then GitHub - enter repo URL - Agent will take care of the rest It will set up the dev and deploym Over 30 new plugins are now available on the Cursor Marketplace.
3 sources discuss meta & llama, including @mattshumer_, @soumithchintala, @percyliang. My mind is blown. @GroqInc is serving LLaMA 3 at over 800 tokens per second! 800. Tokens. Per. Second. This unlocks so many incredible use-cases. It's one thing to see my demo — it's another thi Why do 16k GPU jobs fail? The Llama3 paper has many cool details -- but notably, has a huge infrastructure section that covers how we parallelize, keep things reliable, etc. We hit an overall 90% eff
2 sources discuss openai & gpt, including @mattshumer_, @cursor_ai. I've officially switched away from Cursor. My new stack: - Cline (w/ Sonnet 3.7) for frontend - RepoPrompt (w/ o1 pro) for backend Accuracy and quality are just so much higher when you're not trying It's over. OpenAI just crushed it. We have their o3-level open-source model running on @GroqInc at 500 tokens per second. Watch it build an entire SaaS app in just a few seconds. This is the new
3 sources discuss ai agents, including @databricks, @mattshumer_, @RichardSocher. Data + AI Summit is where data, dev and AI teams go to level up. Whether you're building AI agents, scaling real-time pipelines, or trying to get more projects out of pilot and into production, the 2 Databricks CEO and co-founder @alighodsi joined @technology to discuss the launch of Genie Code, an autonomous AI agent built specifically for data teams. "Just six months ago, we were talking ab
@databricks on ai startup funding. What are the defining traits of companies that put their AI initiatives into production? We looked at actual usage data across 20,000+ customers worldwide to identify the building blocks to a success
The company is currently letting users in the U.K. try out Alexa+ for free via an early access program.
Would you share your medical records with a personal trainer? How about a virtual one? Google, which this week announced it is giving Fitbit's AI health coach the ability to read your medical records, is hoping the answer is yes, following rivals like Amazon, OpenAI, and Microsoft in betting that users are willing to trade their most sensitive data in exchange for more personalized health advice. Starting next month in preview, US Fitbit users will be able to link their medical records to th...
Meta's Llama 3.1 405B is the first open-source model competitive with top closed models. Available for research and commercial use under a permissive license.
1 points • 0 comments on Hacker News
1 points • 0 comments on Hacker News
1 points • 0 comments on Hacker News
1 points • 0 comments on Hacker News
2 points • 0 comments on Hacker News
1 points • 1 comments on Hacker News
1 points • 0 comments on Hacker News
1 points • 0 comments on Hacker News
Windows 11 24H2 update brings deeper Copilot integration with system-level AI assistance, smart file management, and enhanced productivity tools.
92 points • 63 comments on Hacker News
How OpenAI uses chain-of-thought monitoring to study misalignment in internal coding agents—analyzing real-world deployments to detect risks and strengthen AI safety safeguards.
39 points • 2 comments on Hacker News
SD3.5 Large brings improved image quality, better text rendering, and enhanced prompt following. Available under an open license for commercial use.
After compressing models from major AI labs including OpenAI, Meta, DeepSeek and Mistral AI, Multiverse Computing has launched both an app that showcases the capabilities of its compressed models and an API that makes them more widely available.
@ollama on meta & llama. Nemotron 3 Nano 4B is now available to run via Ollama: ollama run nemotron-3-nano:4b Try it with Pi, the minimal agent runtime that powers OpenClaw: ollama launch pi --model nemotron-3-nano:4b Thi Ollama 0.18.1 is here! 🌐 Web search and fetch in OpenClaw Ollama now ships with web search and web fetch plugin for OpenClaw. This allows Ollama's models (local or cloud) to search the web for t
The AI-first code editor Cursor raises Series A funding to expand their team and build next-generation programming tools powered by large language models.
1 points • 0 comments on Hacker News
1 points • 1 comments on Hacker News