Peer announced today it's raised new funding to build its AI-powered metaverse engine and launched its "personal planets" feature.Read More

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn MoreOne goal for an agentic future is for AI agents from different organizations to freely and seamlessly talk to one another. But getting to that point requires interoperability, and these agents may have been built with different LLMs, data frameworks and code.To achieve interoperability, developers of these agents must agree on how they can communicate with each other. This is a challenging task.A group of companies, including Cisco, LangChain, LlamaIndex, Galileo and Glean, have now created AGNTCY, an open-source collective with the goal of creating an industry-standard agent interoperability language. AGNTCY aims to make it easy for any AI agent to communicate and exchange data with another.Uniting AI AgentsJust like when the cloud and the internet came about and accelerated applications and all social interactions at a global scale, we want to build the Internet of Agents that accelerate all of human work at a global scale, said Vijoy Pandey, head of Outshift by Cisco, Ciscos incubation arm, in an interview with VentureBeat.Pandey likened AGNTCY to the advent of the Transmission Control Protocol/Internet Protocol (TCP/IP) and the domain name system (DNS), which helped organize the internet and allowed for interconnections between different computer systems.The way we are thinking about this problem is that the original internet allowed for humans and servers and web farms to all come together, he said. This is the Internet of Agents, and the only way to do that is to make it open and interoperable.Cisco, LangChain and Galileo will act as AGNTCYs core maintainers, with Glean and LlamaIndex as contributors. However, this structure may change as the collective adds more members.Standardizing a fast-moving industryAI agents cannot be islands. To reach their full potential, they must be able to communicate with other agents that lie outside of an enterprises network. This is where interoperability comes in.Setting standards in traditional industries is challenging enough; it becomes even more difficult for technology like AI, where upgrades and model changes occur every few months. However, this is not the first time a standard has been proposed for generative AI.LangChain, one of AGNTCYs core members, has its own protocol for working with agents built on frameworks other than LangChain. The Agent Protocol, launched in November last year, allows LangChain agents to talk to agents created with AutoGen, CrewAI or any other framework.Meanwhile, Anthropic announced its Model Context Protocol (MCP) in November. This protocol aims to standardize how models and AI tools connect to data sources. But while many developers have embraced MCP, its not exactly a standard just yet.Yash Sheth, cofounder of AI evaluation platform Galileo, said standardization is critical.Standardization is needed, in fact, it will drive increased velocity for agentic adoption. Today, teams are building in silos, having to figure out how to develop their own infrastructure components from scratch, Sheth said in an email. Standardization of multi-agentic systems can only happen if these agents powered by non-deterministic models have a strong anchor in measuring and reporting their performance, accuracy and reliability.Sheth admits that making AI agents interoperable can be complex. AGNTCY wants to encourage developers to extend these specs, APIs and tools to suit their needs instead of reinventing the wheel, which will be crucial to achieving standardization.LangChain CEO Harrison Chase said in a separate conversation that creating a standard is not impossible, especially now that its easier to build the agents themselves.Building agents is already possible, and being done. Replit, Klarna, LinkedIn, Uber, Appfolio and many others have all already done this. Agents arent a thing of the future, they are now. Now that we know how to build agents, the next step is to allow them to connect to each other. That is what a standard agent protocol will help enable, Chase said.A platform and a language all at oncePandey envisions AGNTCY as more than just a set of codes for agents. It will also allow customers to discover agents from different developers who run the AGNTCY standard.Customers can stitch together all these agents on the AGNTCY platform so they can discover, compose, deploy and evaluate as they build their workflows, Pandey said.AGNTCY still needs to recruit more AI players to add new agents to the platform and gain momentum as a standard. After all, for something to become an industry standard, there needs to be mass adoption, to prevent the establishment of too many competing standards.Thats where projects like AGNTCY face an uphill battle. Pandey said the collective has been speaking with many other industry players, and they want to get as many viewpoints as possible while developing the platform. That will take time.In the meantime, enterprises continue to experiment and even deploy AI agents. Maybe in the future, these will all be able to speak to each other.Daily insights on business use cases with VB DailyIf you want to impress your boss, VB Daily has you covered. We give you the inside scoop on what companies are doing with generative AI, from regulatory shifts to practical deployments, so you can share insights for maximum ROI.Read our Privacy PolicyThanks for subscribing. Check out more VB newsletters here.An error occured.

Epic Games Store pulls free-to-play Dark and Darker as copyright dispute continuesHowever, both free and premium versions of the RPG remain on SteamImage credit: Ironmace News by Vikki Blake Contributor Published on March 6, 2025 A year after it debuted on Epic Games Store in March 2024, the free-to-play version of Dark and Darker has once again been delisted.Developer Ironmace claims the game has been withdrawn due to its on-going copyright legal dispute with Nexon Games.Both editions of the game - the free-to-play Dark and Dark, and premium Dark and Darker: Legendary Status - remain on Steam. The latter is also still available on Epic Games Store."Epic Games has decided to delist Dark and Darker from the Epic Games Store," Ironmace told players via Discord (thanks, PCGN)."The decision appears to be based on claims made by opposing parties in an ongoing legal dispute. We are currently working to understand the exact reasoning behind this removal."It seems that the removal process is not yet fully complete, as Dark and Darker is no longer listed, but Dark and Darker: Legendary Status is still available for purchase. Please be aware of this potential confusion."In preparation for the possibility that Dark and Darker may no longer be playable on Epic Games, we are working on alternative solutions to ensure that players can continue enjoying the game, Ironmace concluded, before apologizing for any "inconvenienced" caused and promising to will share more details as soon as possible.Dark and Darker launched on Steam in February 2023, but was delisted after Nexon alleged that the game began as its own prototype project P3. Ironmace is composed of former Nexon employees, and Nexon alleged that "stolen" code and assets were used by the team to make Darker and Darker.In March 2023, Ironmace was raided by the Korean police who seized material related to Dark and Darker, but in January 2024, a South Korean court dismissed Nexon's preliminary injunction against Ironmace.

Justin Carter, Contributing EditorMarch 6, 20252 Min ReadImage via Jackbox Games.At a GlanceWith GameLift Streams, Amazon is offering developers a faster and cheaper way to let players stream their games.Amazon's GameLift service now features streaming capabilities that allow developers and publishers to "build and deliver global, direct-to-player game streaming experiences."The streams will enable them to release games directly onto any device with a web browser, including PCs, mobile phones, and SmartTVs. As Amazon notes, this means companies "[avoid] investing millions of dollars in infrastructure and software development to build your own service. Players can start gaming in just a few seconds, without waiting for downloads or installs."Party game developer Jackbox is GameLift Streams' first proponent, and CTO Evan Jacover lauded how the technology easily integrated with its games. "Based on our experiences talking with other streaming platforms, we were anticipating having to do a custom build, but we didnt need to," he said. "We just took our Windows build, uploaded it to Amazon GameLift Streams, and it ran. The whole process was efficient, and the game looked and sounded great."Jackbox's game library has grown enough to require a unified launcher, and Jacover noted that scaling for each individual title would require "a lot of overhead." Gamelift Streams let the developer scale "for our whole catalog, which makes everything much more efficient. Keeping costs down is the key to making streaming work from a business standpoint, so these efficiencies really add up."To further prove GameLift Streams' capabilities, Jackbox launched a SmartTV app. Longtime players can access their library from the app in a "more elegant and streamlined way," and newcomers without a console have an easier way to explore the developer's work.According to Amazon's blog, the GameLift Streams SDK integrates with a game's already existing services, storefronts, and launchers to enable streaming. Developers can use the SDK console to observe active streams and scale the infrastructure across multiple regions on the Amazon Web Services network. "GameLift Streams is the only solution that enables you to upload your game content onto fully-managed GPU instances in the cloud and start streaming in minutes," it reads, "with little or no modification of your code."Read more about:Amazon GamesAbout the AuthorJustin CarterContributing Editor, GameDeveloper.comA Kansas City, MO native, Justin Carter has written for numerous sites including IGN, Polygon, and SyFy Wire. In addition to Game Developer, his writing can be found at io9 over on Gizmodo. Don't ask him about how much gum he's had, because the answer will be more than he's willing to admit.See more from Justin CarterDaily news, dev blogs, and stories from Game Developer straight to your inboxStay UpdatedYou May Also Like

Justin Carter, Contributing EditorMarch 6, 20252 Min ReadImage via Welevel.At a GlanceMunich-based Welevel has raised $5.7 million to develop an unnanounced sandbox game and a suite of AI tools.Independent studio Welevel recently raised $5.7 million in a round of seed funding. The German studio plans to put that money toward growing its staff and developing AI tools made to "streamline game development and create dynamic, personalized gameplay."According to the announcement, these tools "innovate on world-building, NPC behavior, and quest generation, through sophisticated algorithms that make interactions tailored to each player. This system allows for adaptive storytelling where players choices shape the world in unexpected ways, providing a truly dynamic experience."Venture firm Bitkraft, which previously put invested in Sprocket Games and Lightforge Games (among other ventures), led the majority of Welevel's funding.Welevel's debut project game is a triple-A survival game. Founder and CEO Christian Heimerl believed his previous AI experience at past companies allowed him to recognize "the potential it has to empower small, passionate teams to produce games. In the past four years, our team has engineered tools and an AI platform that allow us to not just rival triple-A studios, but surpass them with a living, breathing world."Along with its unannounced game, the studio plans to branch off to "expansive user-generated content and aspirations to expand into other genres and settings."GenAI-ai-aiThe AI tool boom is here, and developers of many sizes are trying to get in on it.In the triple-A sector, Ubisoft, Microsoft, and EA have used generative AI or made tools of their own to aid in development. Ubisoft has its Ghostwriter tool, and EA employed genAI to help with the athlete likenesses of EA Sports College Football 25.More recently, Microsoft unveiled Muse, its latest AI model it promised will help "generate game visuals, controller actions, or both." Xbox head Phil Spencer hyped up its usefulness with the promise of aiding in game preservation and "effectively supporting human creatives."Earlier this week, Xbox subsidiary Activision Blizzard was found allegedly using the technology to promote nonexistent Guitar Hero and Crash Bandicoot games to see if players would like spinoffs from those franchises.Outside the heavy hitters, smaller teams like Echo Chunk and Jam & Tea have used genAI to make puzzle games and mutiplayer RPGs. Like Welevel, both studios are bough into the idea of the technology's ability to reduce the development workload and "revolutionize games."Read more about:FundingGenerative AIAbout the AuthorJustin CarterContributing Editor, GameDeveloper.comA Kansas City, MO native, Justin Carter has written for numerous sites including IGN, Polygon, and SyFy Wire. In addition to Game Developer, his writing can be found at io9 over on Gizmodo. Don't ask him about how much gum he's had, because the answer will be more than he's willing to admit.See more from Justin CarterDaily news, dev blogs, and stories from Game Developer straight to your inboxStay UpdatedYou May Also Like

Yesterday, reviewers agreed: AMD is finally beating Nvidia with its RX 9070 and RX 9070 XT graphics cards, assuming they actually launch with sufficient stock at their $549 and $599 prices.But this mornings launch, while better than recent Nvidia launches in terms of supply, may not have met that bar and AMD would not deny reports that even the best-priced models of these GPUs are capped to specific quantities, after which they may be sold at higher than $549 or $599.When The Verge checked this morning, major US retailers Best Buy and Newegg listed as few as a single model of each GPU at their MSRP, which sold out quickly, of course. Most cards are listed for at least a $100 premium over that MSRP, if not more.Micro Center appears to be doing far better, with as many as five 9070s and five 9070 XTs at MSRP and a website that claims theyre even still available for purchase if you visit one of its 28 US locations in person.What happens next? Its not unusual for each graphics card partner to sell a variety of fancier grades of the same GPU at higher prices with fancier cooling and overclocking but retailers are now suggesting that even the prices of entry-level cards wont hold.We have now learned how the recommended prices, also known as MSRP prices, work for the launch of the AMD Radeon RX 9070 and RX 9070 XT [] they will apply to a limited number of cards, wrote Swedish retailer Inet.se, as documented by VideoCardz.MSRP is capped quantity of a few hundred, so prices will jump once those are sold through, UK retailer Overclockers UK also reportedly wrote.Even Best Buy seemed to suggest that $549 / $599 prices were only temporary. When we visited this morning, the Best Buy website described them like they were on sale, though the company has since removed the save $80 and save $130 you see below.We asked AMD:Can you confirm or deny that the best prices on the best-priced cards are capped in this way?Can you confirm or deny that only a fraction of the best-priced models of 9070 and 9070 XT will be sold at their announced MSRP?Can you confirm or deny that AMD has OKd any board partners or retailers whatsoever to raise their prices beyond MSRP after selling through a certain quantity or certain number of shipments?AMD would not confirm or deny. It did address the larger question of MSRP availability with this statement from chief gaming solutions architect (and gaming marketing boss) Frank Azor, though:It is inaccurate that $549 / $599 MSRP is launch-only pricing. We expect cards to be available from multiple vendors at $549 / $599 (excluding region specific tariffs and / or taxes) based on the work we have done with our AIB partners, and more are coming. At the same time, the AIBs have different premium configurations at higher price points and those will also continue.On X, Azor tweeted that AMD is working to replenish stock in the coming days and weeks, adding: MSRP pricing (excluding region specific tariffs and/or taxes) will continue to be encouraged beyond today so dont despair.See More:

Google Maps is the first app were seeing adopt Android 16s new Live Updates feature, according to Android Authority. Users on Android 16 beta 2.1 can now see the time until their next turn and estimated time to arrival on the status bar of their device.Google introduced the Live Updates feature in its first beta of Android 16 beta release, describing it as a new class of notifications that help users monitor and quickly access important ongoing activities. The feature is similar to Live Activities on iOS, which shows real-time notifications on the lock screen and the Dynamic Island, but until now we havent seen much of what it looks like with real Android apps.As shown by Android Authority, theres also a new option in the Google Maps app that allows you to turn off live info, which will appear on the status bar and lockscreen. Live Updates are still only partially supported in Android 16 beta 2.1, as Android Authority says notifications from Google Maps are still collapsed on the always-on display.Even though Google Maps adoption is still limited, theres plenty of time for Google and other developers to add full support for Live Updates, as Android 16 isnt expected to launch until June.

There is a new Humble Bundle of interest for game developers, the POLYGON Best of Synty GameDev Assets #4 Humble Bundle. The bundle contains assets for both Unreal Engine and Unity (and raw forms, so usable in Godot, Flax, O3DE, etc) in a low polygon style. It is organized into the following tiers:1$ TierPOLYGON Particle FX PackPOLYGON WerewolfSimple OfficeMid TierPOLYGON Dog PackANIMATION Base LocomotionPOLYGON Construction PackSimple People 3Simple TemplesSimple Port30$ TierPOLYGON Dungeon RealmsPOLYGON Kids PackINTERFACE Sci-Fi Soldier HUDPOLYGON Meadow Forest Nature BiomeSimple SpaceSimple Space InteriorsSimple AirportSimple Space Characters$10 VoucherYou can learn more about the POLYGON Best of Synty GameDev Assets #4 Humble Bundle in the video below. Using links on this page help support GFS (and thanks so much if you do!). If you have trouble with the link, try opening it in a new tab and it should work fine.

In todays rapidly evolving digital landscape, the need for accessible, efficient language models is increasingly evident. Traditional large-scale models have advanced natural language understanding and generation considerably, yet they often remain out of reach for many researchers and smaller organizations. High training costs, proprietary restrictions, and a lack of transparency can hinder innovation and limit the development of tailored solutions. With a growing demand for models that balance performance with accessibility, there is a clear call for alternatives that serve both the academic and industrial communities without the typical barriers associated with cutting-edge technology.Introducing AMD InstellaAMD has recently introduced Instella, a family of fully open-source language models featuring 3 billion parameters. Designed as text-only models, these tools offer a balanced alternative in a crowded field, where not every application requires the complexity of larger systems. By releasing Instella openly, AMD provides the community with the opportunity to study, refine, and adapt the model for a range of applicationsfrom academic research to practical, everyday solutions. This initiative is a welcome addition for those who value transparency and collaboration, making advanced natural language processing technology more accessible without compromising on quality.Technical Architecture and Its BenefitsAt the core of Instella is an autoregressive transformer model structured with 36 decoder layers and 32 attention heads. This design supports the processing of lengthy sequencesup to 4,096 tokenswhich enables the model to manage extensive textual contexts and diverse linguistic patterns. With a vocabulary of roughly 50,000 tokens managed by the OLMo tokenizer, Instella is well-suited to interpret and generate text across various domains.The training process behind Instella is equally noteworthy. The model was trained using AMD Instinct MI300X GPUs, emphasizing the synergy between AMDs hardware and software innovations. The multi-stage training approach is divided into several parts:ModelStageTraining Data (Tokens)DescriptionInstella-3B-Stage1Pre-training (Stage 1)4.065 TrillionFirst stage pre-training to develop proficiency in natural language.Instella-3BPre-training (Stage 2)57.575 BillionSecond stage pre-training to further enhance problem solving capabilities.Instella-3B-SFTSFT8.902 Billion (x3 epochs)Supervised Fine-tuning (SFT) to enable instruction-following capabilities.Instella-3B-InstructDPO760 MillionAlignment to human preferences and strengthen chat capabilities with direct preference optimization (DPO).Total:4.15 TrillionAdditional training optimizations, such as FlashAttention-2 for efficient attention computation, Torch Compile for performance acceleration, and Fully Sharded Data Parallelism (FSDP) for resource management, have been employed. These choices ensure that the model not only performs well during training but also operates efficiently when deployed.Performance Metrics and InsightsInstellas performance has been carefully evaluated against several benchmarks. When compared with other open-source models of a similar scale, Instella demonstrates an average improvement of around 8% across multiple standard tests. These evaluations cover tasks ranging from academic problem-solving to reasoning challenges, providing a comprehensive view of its capabilities.The instruction-tuned versions of Instella, such as those refined through supervised fine-tuning and subsequent alignment processes, exhibit a solid performance in interactive tasks. This makes them suitable for applications that require a nuanced understanding of queries and a balanced, context-aware response. In comparisons with models like Llama-3.2-3B, Gemma-2-2B, and Qwen-2.5-3B, Instella holds its own, proving to be a competitive option for those who need a more lightweight yet robust solution. The transparency of the projectevidenced by the open release of model weights, datasets, and training hyperparametersfurther enhances its appeal for those who wish to explore the inner workings of modern language models.ConclusionAMDs release of Instella marks a thoughtful step toward democratizing advanced language modeling technology. The models clear design, balanced training approach, and transparent methodology provide a strong foundation for further research and development. With its autoregressive transformer architecture and carefully curated training pipeline, Instella stands out as a practical and accessible alternative for a wide range of applications.Check outTwitterand dont forget to join our80k+ ML SubReddit. Asif RazzaqWebsite| + postsBioAsif Razzaq is the CEO of Marktechpost Media Inc.. As a visionary entrepreneur and engineer, Asif is committed to harnessing the potential of Artificial Intelligence for social good. His most recent endeavor is the launch of an Artificial Intelligence Media Platform, Marktechpost, which stands out for its in-depth coverage of machine learning and deep learning news that is both technically sound and easily understandable by a wide audience. The platform boasts of over 2 million monthly views, illustrating its popularity among audiences.Asif Razzaqhttps://www.marktechpost.com/author/6flvq/Alibaba Released Babel: An Open Multilingual Large Language Model LLM Serving Over 90% of Global SpeakersAsif Razzaqhttps://www.marktechpost.com/author/6flvq/A Step by Step Guide to Deploy Streamlit App Using Cloudflared, BeautifulSoup, Pandas, Plotly for Real-Time Cryptocurrency Web Scraping and VisualizationAsif Razzaqhttps://www.marktechpost.com/author/6flvq/Qwen Releases QwQ-32B: A 32B Reasoning Model that Achieves Significantly Enhanced Performance in Downstream TaskAsif Razzaqhttps://www.marktechpost.com/author/6flvq/Researchers from FutureHouse and ScienceMachine Introduce BixBench: A Benchmark Designed to Evaluate AI Agents on Real-World Bioinformatics Task Recommended Open-Source AI Platform: IntellAgent is a An Open-Source Multi-Agent Framework to Evaluate Complex Conversational AI System' (Promoted)

Sure Fire GPT-4 Prompting For Better Python Voila Data Viz Results 0 like March 6, 2025Share this postAuthor(s): John Loewen, PhD Originally published on Towards AI. Displaying UN human rights data with Voila, GitHub and myBinder.orgThis member-only story is on us. Upgrade to access all of Medium.Sharing data visualizations from Jupyter Notebooks can be challenging, especially for non-technical users who struggle with running Python code.The combination of GPT-4s coding capabilities and Voilas ability to convert Jupyter Notebooks into interactive dashboards opens up a useful workflow for creating and sharing data visualizations even for users with limited coding experience.By prompting GPT-4 to generate complete, working Python code, you can skip manual setup and jump straight to customizing and deploying interactive data visualizations.With GPT-4 handling the code, your focus can stay on the data.Pairing Voila with GitHub and MyBinder.org allows you to publicly host and share these dashboards with only a few clicks.Let me walk you through using GPT-4 prompts to create a fully functional interactive data visualization from the CIRIGHTS Human Rights dataset, powered by Plotly for visualizations and Voila for sharing the final product through GitHub and MyBinder.org.Before we begin, you need to have Jupyter Notebook and Voila installed. You can install Jupyter from a command (terminal) prompt using pip install jupyter.Once installed, you can create a new notebook within your Jupyter environment, and then add the Voila library:Now that we have the right Read the full blog for free on Medium.Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming asponsor. Published via Towards AITowards AI - Medium Share this post