With access now available to the public, but limited to paying X users, you’ll see the following screen during your sign-up process. Rust has proven to be an ideal choice for building scalable, reliable, and maintainable infrastructure. It offers high performance, a rich ecosystem, and prevents the majority of bugs one would typically find in a distributed system. Given our small team size, infrastructure reliability is crucial, otherwise, maintenance starves innovation. Rust provides us with confidence that any code modification or refactor is likely to produce working programs that will run for months with minimal supervision. At the frontier of deep learning research, reliable infrastructure must be built with the same care as datasets and learning algorithms.
One of its most notable features is its ability to not only understand and process human language but to do so with an added layer of nuance and humor. This unique characteristic is a reflection of the distinct personality of its creators, infusing a more human-like quality into its interactions. For questions that stretch beyond its knowledge base, Grok leverages “real-time access” to info on X (and from Tesla, according to Bloomberg). And, similar to ChatGPT, the model has internet-browsing capabilities, enabling it to search the web for up-to-date information about topics. The xAI API offers developers access to our large language model Grok-1 and our digital assistant Grok. Grok passed the exam with a C (59%), while Claude-2 achieved the same grade (55%), and GPT-4 got a B with 68%.
Particularly interested in improving the robustness of LLMs, reward models, and monitoring systems. Early access to Grok is now available to users who purchased the premium version of X, formerly known as Twitter. Meta will be shutting down Spark AR, its platform of third-party AR tools and content, effective January 14, 2025. Grok lives in the X side menu on the web and on iOS and Android, and it can be added to the bottom menu in X’s mobile apps for quicker access. Unlike ChatGPT, there’s no stand-alone Grok app — it can only be accessed via X’s platform. Our mission is to provide you with great editorial and essential information to make your PC an integral part of your life.
Grok is powered by a large language model (LLM) called Grok-1, in the same way that ChatGPT is powered by GPT-4. An immediate competitor to OpenAI’s ChatGPT, as well as Copilot (formerly Bing Chat), Google Bard, and Anthropic Claude-2, Grok promises to address “spicy” topics such as political views with more humor than its artificially intelligent peers. Pious evangelists of the billionaire refer to its USP as being “anti-woke” or “based”, which translates to “less left-wing in my non-left-wing opinion” for those fortunate souls not chronically online. On these benchmarks, Grok-1 displayed strong results, surpassing all other models in its compute class, including ChatGPT-3.5 and Inflection-1.
- The decision to initially limit access is a strategic one, indicative of a careful and thoughtful approach to its rollout.
- Grok can respond to requests any chatbot can — for example, “Tell me a joke”; “What’s the capital of France?
- We evaluated the Grok-2 models across a series of academic benchmarks that included reasoning, reading comprehension, math, science, and coding.
- Human-guided feedback, or reinforcement learning from human feedback (RLHF), is the way most AI-powered chatbots are fine-tuned these days.
After all, Musk has previously complained that ChatGPT is too «woke,» and said that training AI to be politically correct is «deadly.» Grok AI isn’t merely a new tool; it represents a shift in how we interact with technology. Its launch is a testament to the strides we’re making in AI communication, and learning how to use Grok AI could very well be your stepping stone into a future where AI is not just a tool but a partner. Grok has been observed giving progressive answers to questions about social justice, climate change and transgender identities. In fact, one researcher found its responses on the whole to be left-wing and libertarian — even more so than ChatGPT’s.
Plus, it can be added to the bottom menu on X’s mobile apps for quick access. Grok AI, a recent development by xAI, marks a significant evolution in the field of conversational artificial intelligence. This new AI system stands out from its predecessors in several key ways.
Get the industry’s biggest tech news
When I asked Grok «What do you think of ChatGPT?» in «regular» mode, the chatbot responded with «I think ChatGPT is a powerful tool that can help users in various tasks.» These unique selling points showcase Grok AI’s commitment to providing a rich, dynamic, and flexible user experience. By leveraging these strengths, users can enjoy a conversational AI that is not only informative and up-to-date but also engaging and adaptable to a wide range of conversational needs and preferences. As of now, Grok AI is in its beta testing phase, with access limited to a select group of users. This restricted availability has significantly heightened the curiosity and interest surrounding it.
Watch: What is ChatGPT, and should we be afraid of AI chatbots?
You also need to fork over $16 per month — $168 per year — for an X Premium+ plan. Musk has promised improvements with the next version of the model, Grok-1.5, set to arrive later this year. The main difference (in theory) is that Grok will be more fun, and less cautious. Here’s how to sign up for the Grok waitlist, and begin using Musk’s new AI platform as soon as possible. If you share our optimism and want to contribute to our mission, apply to join the team here. Please leave your email address below if you would like to get notified when you have been given access.
This experiment served as a “real-life” test on a dataset our model was never explicitly tuned for. Grok has real-time access to info via the X platform, which is a massive advantage over other models. Exactly how helpful Grok will be is unclear, as most current users seem more interested in its artificially snarky «how do you do, fellow kids» tone than how accurate its answers are. A preview from a Tesla employee earlier this week was less than reassuring, providing information that was both incorrect and vague to the point of uselessness.
Is open source AI possible, let alone the future? Find out at TechCrunch Disrupt 2024
X Premium+ is the highest-priced subscription on X, as it removes all the ads in the For You and Following feeds. In addition, Premium+ introduces a hub where users can get paid to post and offer fans subscriptions, and Premium+ users have their replies boosted the most in X’s rankings. We provide a summary of the important technical details of Grok-1 in the model card. To understand the capability improvements we made with Grok-1, we have conducted a series of evaluations using a few standard machine learning benchmarks designed to measure math and reasoning abilities. The engine powering Grok is Grok-1, our frontier LLM, which we developed over the last four months.
Engineering at xAI
It is only surpassed by models that were trained with a significantly larger amount of training data and compute resources like GPT-4. This showcases the rapid progress we are making at xAI in training LLMs with exceptional efficiency. Since announcing Grok-1 in November 2023, xAI has been moving at an extraordinary pace, driven by a small team with the highest talent density. We have introduced Grok-2, positioning us at the forefront of AI development. Our focus is on advancing dogecoins 60% growth over q2 proves it should be taken significantly core reasoning capabilities with our new compute cluster. We are looking for individuals to join our small, focused team dedicated to building the most impactful innovations for the future of humanity.
We want Grok to serve as a powerful research assistant for anyone, helping them to quickly access relevant information, process data, and come up with new ideas. One version is called «Regular Mode,» which appears to spit out matter-of-fact responses to queries. The other is «Fun Mode,» which will generate responses in a humorous, sarcastic tone. Both the $16-a-month plan and the $168 annual plan give you early access to Grok. Keep your eyes peeled for updates on how to use Grok AI as it evolves and becomes more widely available. Whether you’re a tech aficionado or just someone who loves a good chat, Grok AI is set to redefine your digital interaction experience.
Amanda Yeo is an Assistant Editor at Mashable, covering entertainment, culture, tech, science, and social good. Based in Australia, she writes about everything from video games and K-pop to movies and gadgets. And asking Grok about its own personal thoughts can lead to drastically different answers. For instance, I asked Grok «What are your AI predictions for 2024?» Both versions spit out similar answers like AI-driven healthcare and automating business operations. Depending on what you ask, different versions of Grok can generate their own unique responses. Once billing is complete, you should now be able to access the AI chatbot through your feed.
This phased deployment aims to ensure that when Grok AI becomes more widely available, it will offer a highly polished and efficient user experience. The focus during the beta phase is on meticulously refining the AI’s capabilities, fine-tuning its responses, and ensuring its reliability across various types of user interactions. This period of exclusivity allows the developers to gather valuable feedback from initial users, address any potential issues, and make improvements based on real-world usage. It is engineered to engage with users more interactively and dynamically, offering assistance in a wide range of activities. Whether it’s handling everyday tasks or navigating more complex scenarios, it is adept at providing support.
This feature enables Grok to provide answers with up-to-the-minute information, making it a standout in the world of chatbots. Unlike Bard and ChatGPT, which often offer vague and outdated responses due to their training data limitations, Grok leverages recent headlines to craft its answers. After announcing xAI, we trained a prototype LLM (Grok-0) with 33 billion parameters. This early model approaches LLaMA 2 (70B) capabilities on standard LM benchmarks but uses only half of its training resources. We are also releasing Grok-2 and Grok-2 mini to developers through our new enterprise API platform later this month.