Categories
News

Open-Source Mistral AI Model now available on IBM watsonx

Mistral AI Model on IBM watsonx

IBM has taken a bold step by incorporating an advanced AI model known as Mixtral-8x7B, which comes from the innovative minds at Mistral AI. This is a big deal because it means you now have access to a broader range of AI models to choose from, allowing you to tailor your AI solutions to fit your unique business needs perfectly.

The Mixtral-8x7B model is a powerhouse in the realm of large language models (LLMs). It’s designed to process data at lightning speeds, boasting a 50% increase in data throughput. This is a significant advantage for any business that relies on quick and efficient data analysis. Imagine reducing potential latency by up to 75%—that’s the kind of speed we’re talking about.

But speed isn’t the only thing this model has going for it. The Mixtral-8x7B is also incredibly efficient, thanks to a process called quantization. This technique shrinks the model’s size and reduces its memory requirements, which can lead to cost savings and lower energy consumption. And the best part? It does all this without compromising on its ability to handle complex data sets.

Mistral AI Model on watsonx

IBM’s strategy is all about giving you options. With a diverse range of AI models on the Watsonx platform, you can pick and choose the tools that best fit your business operations. The Mixtral-8x7B model is a testament to this approach, offering versatility for a variety of business applications.  Collaboration is at the heart of IBM’s model development. By working with other AI industry leaders like Meta and Hugging Face, IBM ensures that its Watsonx.ai model catalog is stocked with the latest and greatest in AI technology. This means you’re always getting access to cutting-edge tools.

The Mixtral-8x7B model isn’t just fast and efficient; it’s also smart. It uses advanced techniques like Sparse modeling and Mixture-of-Experts to optimize data processing and analysis. These methods help the model manage vast amounts of information with precision, making it an invaluable asset for businesses drowning in data. IBM’s global perspective is evident in its recent addition of ELYZA-japanese-Llama-2-7b, a Japanese LLM, to the Watsonx platform. This move shows IBM’s dedication to catering to a wide range of business needs and use cases across different languages and regions.

Looking ahead, IBM isn’t stopping here. The company plans to keep integrating third-party models into Watsonx, constantly enhancing the platform’s capabilities. This means you’ll have an ever-expanding toolkit of AI resources at your disposal. So, what does IBM’s integration of the Mixtral-8x7B model into Watsonx mean for you? It signifies a major leap forward in the company’s AI offerings. With a focus on increased efficiency, a robust multi-model strategy, and a commitment to collaboration, IBM is well-equipped to help you leverage AI for a competitive edge in your industry. Whether you’re looking to innovate, scale, or simply stay ahead of the curve, IBM’s Watsonx platform is becoming an increasingly valuable ally in the fast-paced world of enterprise AI. Here are some other articles you may find of interest on the subject of  Mixtral and IBM watsonx :

Filed Under: Technology News, Top News





Latest timeswonderful Deals

Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, timeswonderful may earn an affiliate commission. Learn about our Disclosure Policy.

Categories
News

MinIO open-source object storage data solution with high performance and S3 compatibility

MinIO open-source object storage solution

If you are looking for a high performance open source storage solution you might be interested in MinIO. Offering S3 compatibility and allowing users to host their data on their own servers. It provides an alternative to AWS S3, addressing concerns about proprietary software risks and potential performance issues. MinIO supports features such as buckets, access policies, versioning, and object locking. It can be installed on a user’s own server or through a cloud platform service, which can handle installation, backup, updates, and maintenance.

The platform’s user interface includes options for dark and light modes, and it guides users through creating buckets and managing data. MinIO also offers the ability to create public or private buckets, version control, and access management through user and service accounts. Advanced features include event-based actions, monitoring, metrics, and tiering for automated data management. The platform is scalable, supporting site replication and cluster expansion.

MinIO is a heavyweight in the world of data storage, offering a level of performance that can go toe-to-toe with industry giants like AWS S3. But what sets it apart is its open-source nature, giving you the reins to steer your data management in any direction you choose. This is especially valuable if you’re looking to spread your storage solutions across different platforms or if you’re planning to migrate away from AWS S3.

One of the standout features of MinIO is its seamless compatibility with Amazon S3 APIs. This means you can easily integrate it with a plethora of services and tools you’re already using, making the transition smooth and hassle-free. If you’re considering diversifying your storage options or moving away from a sole reliance on AWS S3, MinIO’s compatibility is a significant advantage.

MinIO Data Object Storage

Here are some other articles you may find of interest on the subject of coding

When it comes to handling your data, MinIO doesn’t play around. You can set it up on your own infrastructure, or opt for a cloud service that takes care of the installation and upkeep for you. This flexibility allows you to choose the path that best aligns with your organization’s needs and goals. The user interface of MinIO is a breath of fresh air, designed to be straightforward and user-friendly. It doesn’t matter if you’re a fan of dark or light mode; the interface has got you covered with all the essential tools you need. You can create buckets, manage versions, lock objects, and set quotas—all with a few clicks.

But what about keeping your data safe? MinIO has you covered there, too. You can create both public and private buckets, ensuring that your data is accessible only to those you permit. The platform also comes equipped with robust user and service account management tools, so you can set precise access policies and maintain tight control over who can modify or view your data.

MinIO is Built to Scale

Efficiency is key in any operation, and MinIO understands that. It offers event-based actions that can automate your workflows, boosting your operational efficiency. Plus, with its monitoring tools, you’ll get valuable insights into how your storage is being used and how it’s performing, enabling you to make smart, data-driven decisions.

As your organization grows, so do your storage needs. MinIO is built to scale right alongside you. It supports site replication and cluster expansion, which means you can increase your storage capacity whenever you need to. And you won’t have to worry about performance taking a hit as you grow; MinIO is engineered to provide rapid data access, no matter how large your storage demands become.

MinIO is more than just a storage solution; it’s a powerful tool that offers you the flexibility and control you need to manage your data effectively. With its S3 compatibility, user-friendly interface, and advanced features for scalability and automation, MinIO is an excellent choice for any organization looking to step up its data storage game. Whether you’re searching for an alternative to AWS S3 or you simply desire more control over your storage infrastructure, MinIO has the capabilities to meet—and exceed—your expectations.

So, if you’re ready to take the next step in data storage and management, consider MinIO. It’s not just about storing data; it’s about empowering you to manage, secure, and scale your data storage in a way that aligns with your organization’s unique needs. With MinIO, you’re not just choosing a storage solution; you’re choosing a partner that grows with you, ensuring that your data is always in the right hands—yours.

Filed Under: Technology News, Top News





Latest timeswonderful Deals

Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, timeswonderful may earn an affiliate commission. Learn about our Disclosure Policy.

Categories
News

Mistral-NEXT new open-source model fully tested

Mistral-NEXT new open-source model fully tested

Unfortunately we might be coming to the end of the line for open source AI models from Mistral AI after the company quietly released their latest Mistral Large and speculations  that perhaps they would be moving from an open source release system. Mistral AI has made a name for itself by creating open-source AI models such as Mistral and Mixtral. The latest Mistral-NEXT model was discovered despite no official announcement and its capabilities have now been tested against various tasks, including coding, logic and reasoning, and content generation.

This new AI large language model has caught the attention of many due to its impressive abilities in a range of tasks. It’s the latest in a series of open-source AI models that have been making waves in the tech world.  Starting with the basics, Mistral-NEXT has shown that it can handle simple computational tasks with ease. This means it can be a reliable tool for performing basic operations, which is great news for those looking for a dependable AI to assist with straightforward calculations.

However, when it comes to more complex tasks like coding, the model’s performance is mixed. For instance, it can write a Python script for a game, but the code isn’t perfect. It understands the language and the mechanics of the game, but to get the best results, a human touch is needed to refine the work. The model’s ability to solve problems using logic and reasoning is one of its standout features. It can work through a variety of challenges accurately, showing that it has a strong foundation for tackling these kinds of tasks.

Mistral-NEXT performance tested

Content generation is another area where Mistral-NEXT has proven itself to be capable. However, it’s important to note that when creating content, especially if it’s sensitive or needs to be in a specific format like JSON, human oversight is still necessary to ensure the output is of high quality and appropriate.

Here are some other articles you may find of interest on the subject of Mistral AI :

When we compare Mistral-NEXT to the more advanced GPT-4, it holds its own, particularly in logic and reasoning. But there are areas where GPT-4 might have the upper hand, possibly because it has been trained on a larger dataset or uses more complex algorithms. This comparison is important as it helps us understand where Mistral-NEXT stands in the current AI landscape and what it might achieve in the future.

The AI community is watching closely to see if Mistral-NEXT will be made available as an open-source model on platforms like Hugging Face. The decision to open-source a model like this can have a big impact. It can lead to wider adoption and improvements as the community gets involved, contributing to the model’s development and enhancing its capabilities through collaboration.

The Mistral-NEXT model has shown a lot of promise in its performance tests. It’s particularly adept at logic and reasoning tasks. However, there’s still room for it to grow and improve, especially when compared to more advanced models like GPT-4. The AI field is looking forward to seeing what the future holds for Mistral-NEXT. If it becomes open-source, it could lead to a wave of collaborative innovation and significant progress in the field of artificial intelligence.

Filed Under: Technology News, Top News





Latest timeswonderful Deals

Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, timeswonderful may earn an affiliate commission. Learn about our Disclosure Policy.

Categories
News

Apple releases MGIE open-source AI image editor

Apple releases MGIE open-source AI image editor

In the realm of digital image editing, Apple’s recent unveiling of the Multimodal Large Language Model-Guided Image Editing (MGIE) system marks a significant milestone. This cutting-edge AI tool leverages the capabilities of large language models to interpret and execute complex, instruction-based image modifications, offering users an unprecedented level of control and flexibility. MGIE’s innovative approach combines the power of text and visual inputs to facilitate Photoshop-style adjustments, global photo enhancements, and precise local edits with remarkable efficiency.

Apple MGIE

The development of MGIE embodies Apple’s commitment to pushing the boundaries of technology and creativity, providing a platform that not only simplifies sophisticated editing tasks but also encourages collaboration and innovation within the open-source community. By integrating multimodal learning techniques, MGIE significantly improves upon previous image editing systems, enabling more expressive and accurate interpretations of user instructions. Providing open source competition to the likes of Midjourney and OpenAI’s DallE 3.

Open source image editor

In recent years, the intersection of artificial intelligence and creative tools has led to revolutionary advances in how we interact with digital media. Apple’s introduction of the MGIE system stands as a testament to this ongoing transformation, setting a new standard for AI-powered creativity.

MGIE (MLLM-Guided Image Editing), an open-source AI model developed in collaboration with University of California researchers. This model, highlighted for its ability to perform intricate image manipulations based on natural language instructions, leverages multimodal large language models (MLLMs) to accurately interpret user requests. MGIE enables a wide range of edits, from global photo enhancements like adjusting brightness and contrast to local modifications and Photoshop-style alterations such as cropping, resizing, and adding filters.

iOS 18

Its capability to understand and execute commands like making a pizza look healthier or altering the focus in a photo showcases its advanced common sense reasoning and pixel-level manipulation skills. MGIE’s development, shared at the International Conference on Learning Representations (ICLR) 2024 and available on GitHub, signifies a major leap forward in AI research for Apple, following closely on the heels of other significant AI projects and the anticipation of generative AI features in iOS 18.

Apple MGIE AI image editor

MGIE represents a bridge between advanced AI capabilities and user-friendly image editing, enabling a plethora of modifications ranging from global photo enhancements like brightness, contrast, and sharpness adjustments to more focused local edits that can alter the shape, size, color, or texture of specific image elements. Furthermore, it excels in Photoshop-style operations, including cropping, resizing, rotating, and applying various filters, offering users an unprecedented level of control over their digital environments.

Multimodal Large Language Model-Guided Image Editing

One of the most remarkable aspects of MGIE is its common-sense reasoning ability, which allows it to perform tasks such as adding vegetable toppings to a pizza to make it appear healthier or enhancing a photo’s contrast to simulate additional light. This level of intuitive operation paves the way for more creative and personalized image editing, pushing the boundaries of what can be achieved with AI technology.

The collaboration with the University of California and the presentation of MGIE at the International Conference on Learning Representations (ICLR) 2024 mark a milestone in Apple’s AI research endeavors. Available on GitHub, MGIE invites further exploration and development, providing access to its code, data, and pre-trained models to the broader scientific and creative communities.

AI image generation and manipulation research

This initiative is part of Apple’s broader commitment to AI research, as evidenced by its recent achievements in deploying large language models on iPhones and other devices with limited memory. The development of an “Apple GPT” rival and the “Ajax” framework for large language models underscore the company’s dedication to advancing AI technology. Furthermore, the anticipation of generative AI features in iOS 18, including an enhanced version of Siri with ChatGPT-like functionality, signals a significant shift in how AI will integrate into everyday devices, potentially marking the “biggest” software update in iPhone history according to industry analysts.

MGIE is not just a tool but a harbinger of the future of digital creativity, blending the lines between technological innovation and artistic expression. Its development and open-source release underscore Apple’s vision of a world where technology serves not only to enhance productivity but also to foster creativity and personal expression through intuitive, accessible, and powerful tools. As MGIE evolves, it is set to redefine the landscape of image editing, making advanced AI-driven image manipulation accessible to a wider audience and encouraging a new era of digital artistry.

Filed Under: Apple, Top News





Latest timeswonderful Deals

Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, timeswonderful may earn an affiliate commission. Learn about our Disclosure Policy.

Categories
News

Eagle-7B open-source AI model uses RWKV-v5 architecture

Eagle-7B open-source AI model uses RWKV-v5 architecture

A new open source AI model has emerged that could reshape the way we think about language processing. The Eagle-7B model, a brainchild of RWKV and supported by the Linux Foundation, is making waves with its unique approach to handling language. Unlike the Transformer models that currently dominate the field, Eagle-7B is built on a recurrent neural network (RNN) framework, specifically the RWKV-v5 architecture. This model is not just another iteration in AI technology; it’s a step forward that promises to make language processing faster and more cost-effective.

One of the most striking aspects of Eagle-7B is its commitment to energy efficiency. In a world where the environmental impact of technology is under scrutiny, Eagle-7B stands out for its low energy consumption during training. This makes it one of the most eco-friendly options among large language models (LLMs), a critical consideration for sustainable development in AI.

But Eagle-7B’s prowess doesn’t stop at being green. It’s also a polyglot’s dream, trained on an extensive dataset that includes over 1.1 trillion tokens across more than 100 languages. This extensive training has equipped Eagle-7B to handle multilingual tasks with ease, often performing on par with or even better than much larger models like Falcon 1.5 trillion and Llama 2 trillion.

Eagle-7B – RWKV-v5

Here are some other articles you may find of interest on the subject of AI models

The technical innovation of Eagle-7B doesn’t end with its linguistic abilities. The model’s hybrid architecture, which combines RNNs with temporal convolutional networks (TCNs), brings a host of benefits. Users can expect faster inference times, less memory usage, and the ability to process sequences of indefinite length. These features make Eagle-7B not just a theoretical marvel but a practical tool that can be applied to a wide range of real-world scenarios.

Accessibility is another cornerstone of the Eagle-7B model. Thanks to its open-source licensing under Apache 2, the model fosters collaboration within the AI community, encouraging researchers and developers to build upon its foundation. Eagle-7B is readily available on platforms like Hugging Face, which means integrating it into your projects is a straightforward process.

Features of the Eagle-7B  AI model include :

  • Built on the RWKV-v5 architecture
    (a linear transformer with 10-100x+ lower inference cost)
  • Ranks as the world’s greenest 7B model (per token)
  • Trained on 1.1 Trillion Tokens across 100+ languages
  • Outperforms all 7B class models in multi-lingual benchmarks
  • Approaches Falcon (1.5T), LLaMA2 (2T), Mistral (>2T?) level of performance in English evals
  • Trade blows with MPT-7B (1T) in English evals
  • All while being an “Attention-Free Transformer”
  • Is a foundation model, with a very small instruct tune – further fine-tuning is required for various use cases!
  • We are releasing RWKV-v5 Eagle 7B, licensed as Apache 2.0 license, under the Linux Foundation, and can be used personally or commercially without restrictions
  • Download from Huggingface, and use it anywhere (even locally)
  • Use our reference pip inference package, or any other community inference options (Desktop App, RWKV.cpp, etc)
  • Fine-tune using our Infctx trainer

d continuous performance improvements, ensuring that it remains adaptable and relevant for various applications. Its scalability is a testament to its potential, as it can be integrated into larger and more complex systems, opening up a world of possibilities for future advancements.

The launch of Eagle-7B marks a significant moment in the development of neural networks and AI. It challenges the prevailing Transformer-based models and breathes new life into the potential of RNNs. This model shows that with the right data and training, RNNs can achieve top-tier performance.

Eagle-7B is more than just a new tool in the AI arsenal; it represents the ongoing quest for innovation within the field of neural networks. With its unique combination of RNN and TCN technology, dedication to energy efficiency, multilingual capabilities, and open-source ethos, Eagle-7B is set to play a pivotal role in the AI landscape. As we continue to explore and expand the boundaries of AI technology, keep an eye on how Eagle-7B transforms the standards of language processing.

Image Credit : RWKV

Filed Under: Technology News, Top News





Latest timeswonderful Deals

Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, timeswonderful may earn an affiliate commission. Learn about our Disclosure Policy.

Categories
News

OpenCopilot free and open-source AI Copilot for SaaS products

Open Copilot open-source AI assistant

If you would like a little help adding extra functionality to your SaaS product with the assistance of an artificial intelligent. OpenCopilot is an AI copilot and assistant that features a, cloud-based dashboard that you can access from anywhere, at any time. Enabling your SaaS product to harness the power of AI.

OpenCopilot is a free and open-source tool that allows you to create AI copilots for your SaaS product. The copilot can communicate with your underlying APIs and execute calls as needed. We are open-source under MIT license and also support OpenAPI standards.

The beauty of this is that you don’t have to worry about complicated setups or configurations. You can dive right into your tasks with minimal fuss. The dashboard itself is designed to be user-friendly, so you can hit the ground running without needing to spend hours learning how to use it.

The planning engine of the AI has been significantly enhanced to offer better task management capabilities. This means that the AI can help you juggle multiple projects or focus on urgent tasks with a level of precision that wasn’t possible before. It’s like having a personal assistant who’s always one step ahead, ready to adapt to whatever your project throws at you.

OpenCopilot AI assistant

One of the standout features of OpenCopilot is the ability to keep track of and review your chat conversations with the AI. This is incredibly useful because it allows you to look back on previous interactions, which can be helpful when you need to pick up where you left off or evaluate the guidance you’ve received.

Here are some other articles you may find of interest on the subject of AI assistants.

The AI copilot is powered by advanced language models, which means it can understand your instructions and carry out API interactions with ease. Whether you need help with editing, debugging, or creating content, all you have to do is tell the AI what you need, and it will handle the rest. It’s like having an expert in your corner, ready to assist with a variety of tasks.

For those who need a tailored experience, the tool offers a range of deployment and customization options. It doesn’t matter if you’re flying solo or working as part of a team; you can set up the AI copilot to fit your project’s specific needs. And if you ever find yourself stuck, there’s a wealth of tutorials and documentation available to guide you through the process. Self-managing installations is straightforward, with clear instructions and defined prerequisites. This means you can get your AI copilot up and running quickly, providing you with the support you need to tackle your projects head-on.

Benefits of integrating and AI Copilot into your SaaS product

Integrating an AI Copilot into a Software as a Service (SaaS) product can offer several benefits, enhancing both user experience and operational efficiency:

  • Enhanced User Experience: AI can provide personalized recommendations and support, improving user engagement and satisfaction. By understanding user preferences and behavior, the AI can tailor the user interface, suggest relevant functionalities, or offer help in real-time.
  • Increased Efficiency: AI can automate routine tasks, such as data entry, analysis, or customer support queries. This not only speeds up processes but also allows human staff to focus on more complex, value-added activities.
  • Data-Driven Insights: AI’s ability to analyze large datasets can uncover trends and patterns that might be missed by human analysis. These insights can inform business decisions, from product development to marketing strategies.
  • Improved Accuracy: AI algorithms, especially in areas like data processing or predictive analysis, can operate with a high degree of accuracy, reducing the likelihood of human error.
  • Scalability: An AI copilot can easily handle scaling up operations, managing increased workloads or user numbers more efficiently than a human team.
  • Cost Reduction: Over time, the automation and efficiency gains provided by AI can lead to significant cost savings, as less human intervention is required for many processes.
  • Continuous Improvement: AI systems can learn from interactions and feedback, continuously improving their performance and the service they provide.
  • Competitive Advantage: Offering advanced AI capabilities can distinguish a SaaS product in the market, attracting customers looking for cutting-edge solutions.
  • Customization and Flexibility: AI can adapt to different user needs and scenarios, providing a more flexible and customized service.
  • Enhanced Security: AI can be used to monitor and detect security threats in real-time, offering an additional layer of protection for the SaaS platform and its users.
  • Accessibility and Inclusivity: AI can offer features like language translation, voice recognition, and personalized interfaces, making the SaaS product more accessible to a diverse user base.

Beyond the tool itself, there’s a whole ecosystem of resources at your disposal. You can join a private Discord community to share ideas, learn from others, and get support from peers who are also using the AI copilot. If you’re looking for more advanced features, you can explore various AI tool subscriptions. These platforms also offer a chance to network and potentially create valuable partnerships.

For businesses looking to scale, there are professional consulting services available to help integrate AI solutions into your growth strategy. With these resources, you have everything you need to harness the power of AI for your projects.

The open-source AI copilot tool is more than just an assistant; it’s a powerful ally in your quest for productivity and efficiency. Whether your field is software development, content management, or innovation, this tool is ready to help you achieve your goals. Dive into the features of the AI copilot and consider how it could revolutionize your products and workflows. The future of project management is here, and it’s powered by the incredible capabilities of AI.

Filed Under: Guides, Top News





Latest timeswonderful Deals

Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, timeswonderful may earn an affiliate commission. Learn about our Disclosure Policy.

Categories
News

Microsoft Radius open-source application platform for the cloud

Microsoft Radius

In the ever-evolving world of software development, the cloud has become a central stage where modern applications are brought to life. At the recent Microsoft Ignite conference Microsoft reveal more details about its new Azure’s Radius project specifically designed to guide developers through the intricacies of cloud-native application development. This platform has been created to make the process of building and deploying applications that are scalable and perform flawlessly across different cloud environments much more manageable.

The Microsoft Radius open source project provides a robust and flexible framework that enhances the way developers create cloud-native applications. It arms them with a suite of tools to build applications that are not just resilient and easy to manage but also highly observable. With features like service discovery, load balancing, and automated recovery, Radius ensures that applications are not only ready for today’s demands but also equipped to adapt to future changes in cloud technology.

One of the core principles of Radius is the separation of application code from the underlying infrastructure. This separation allows developers to focus on writing code that delivers real business value without getting bogged down by the complexities of infrastructure management. This clear distinction improves maintainability and speeds up the development cycle, leading to faster iterations and quicker deployments.

Microsoft Radius

Radius also brings a new dimension to platform engineering by providing tools that help create self-service platforms. These platforms empower developers to deploy and manage applications with less dependence on operations teams. Additionally, Radius promotes a culture of collaboration between developers, operators, and architects, creating a shared space where insights and responsibilities are exchanged freely.

Here are some other articles you may find of interest on the subject of Microsoft applications and services :

A key feature of Radius is the introduction of “recipes” that standardize the deployment of infrastructure components. These recipes act as blueprints, ensuring that environments are consistent and deployments can be repeated with precision. This standardization is crucial for reducing errors and ensuring that applications behave predictably from one development stage to the next.

Open Source

Embracing the open-source ethos, Radius supports cloud-neutral development practices, allowing for deployments across various cloud platforms, including on-premises setups. This approach protects against vendor lock-in and encourages contributions from the community, bringing together a wealth of developer expertise and ongoing enhancements to the project.

Security and Compliance

Security and compliance are non-negotiable in the digital world, and Radius takes these aspects seriously. The project is built with these best practices in mind, ensuring that applications are not only agile but also secure. By integrating these practices early in the development process, Radius helps mitigate risks and strengthens applications against potential security threats.

Incremental adoption

Designed for incremental adoption, Radius can be integrated with existing cloud-native initiatives. This flexible approach allows developers to start small and expand their use of the project at a comfortable pace, making the transition to new technologies smoother.

Another innovative feature of Radius is the “application graph,” which provides a visual representation of an application’s architecture and the connections between its components. This tool is invaluable for understanding the structure of an application and identifying areas that could benefit from improvement.

Looking ahead, the Radius project is on track to be submitted to the Cloud Native Computing Foundation (CNCF) for broader adoption and governance. This strategic move will align Radius with industry best practices and standards and place it under the guidance of a respected governing body.

Microsoft Azure’s Radius project is a significant step forward in the field of cloud-native application development. It addresses the complexities of infrastructure management, upholds best practices, and fosters a collaborative environment. Radius equips developers with the necessary tools and principles to create scalable, resilient, and secure applications more easily. Whether deploying to a single cloud or navigating a multi-cloud landscape, Radius stands as a reliable partner, offering the essential components for success in the cloud-native space. With Radius, the path to efficient and effective cloud-native development is clearer, allowing developers to focus on innovation and delivering value through their applications.

Filed Under: Technology News, Top News





Latest timeswonderful Deals

Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, timeswonderful may earn an affiliate commission. Learn about our Disclosure Policy.

Categories
News

Perplexity Lab pplx-api API for open-source LLMs

Perplexity API for open-source LLMs

Perplexity Labs has recently introduced a new, fast, and efficient API for open-source Large Language Models (LLMs) known as pplx-api. This innovative tool is designed to provide quick access to various open-source LLMs, including Mistral 7B, Llama2 13B, Code Llama 34B, and Llama2 70B. The introduction of pplx-api marks a significant milestone in the field of AI, offering a one-stop-shop for open-source LLMs.

One of the key features of pplx-api is its ease of use for developers. The API is user-friendly, allowing developers to integrate these models into their projects with ease using a familiar REST API. This ease of use eliminates the need for deep knowledge of C++/CUDA or access to GPUs, making it accessible to a wider range of developers.

Perplexity Lab pplx-api

The pplx-api also boasts a fast inference system. The efficiency of the inference system is remarkable, offering up to 2.9x lower latency than Replicate and 3.1x lower latency than Anyscale. In tests, pplx-api achieved up to 2.03x faster overall latency compared to Text Generation Inference (TGI), and up to 2.62x faster initial response latency. The API is also capable of processing tokens up to 2x faster compared to TGI. This speed and efficiency make pplx-api a powerful tool for developers working with LLMs.

Benefits of the pplx-api

  • Ease of use: developers can use state-of-the-art open-source models off-the-shelf and get started within minutes with a familiar REST API.

  • Blazing fast inference:  thoughtfully designed inference system is efficient and achieves up to 2.9x lower latency than Replicate and 3.1x lower latency than Anyscale.

  • Battle tested infrastructure: pplx-api is proven to be reliable, serving production-level traffic in both Perplexity answer engine and  Labs playground.

  • One-stop shop for open-source LLMs: Perplexity Labs is dedicated to adding new open-source models as they arrive. For example, we added Llama and Mistral m

The infrastructure of pplx-api is reliable and battle-tested. It has been proven reliable in serving production-level traffic in both Perplexity’s answer engine and Labs playground. The infrastructure combines state-of-the-art software and hardware, including AWS p4d instances powered by NVIDIA A100 GPUs and NVIDIA’s TensorRT-LLM. This robust infrastructure makes pplx-api one of the fastest Llama and Mistral APIs commercially available.

API for open-source LLMs

The pplx-api is currently in public beta and is free for users with a Perplexity Pro subscription. This availability allows a wider range of users to test and provide feedback on the API, helping Perplexity Labs to continually improve and refine the tool. The API is also cost-efficient for LLM deployment and inference. It has already resulted in significant cost savings for Perplexity, reducing costs by approximately $0.62M/year for a single feature. This cost efficiency makes pplx-api a valuable tool for both casual and commercial use.

The team at Perplexity is committed to adding new open-source models as they become available, ensuring that pplx-api remains a comprehensive resource for open-source LLMs. The API is also used to power Perplexity Labs, a model playground serving various open-source models. The introduction of pplx-api by Perplexity Labs represents a significant advancement in the field of AI. Its ease of use, fast inference system, reliable infrastructure, and cost efficiency make it a powerful tool for developers working with open-source LLMs. As the API continues to evolve and improve, it is expected to become an even more valuable resource for the AI community.

In the near future, pplx-api will support:

  • Custom Perplexity LLMs and other open-source LLMs.

  • Custom Perplexity embeddings and open-source embeddings.

  • Dedicated API pricing structure with general access after public beta is phased out.

  • Perplexity RAG-LLM API with grounding for facts and citations.

How to access pplx-api

You can access the pplx-api REST API using HTTPS requests. Authenticating into pplx-api involves the following steps:

1. Generate an API key through the Perplexity Account Settings Page. The API key is a long-lived access token that can be used until it is manually refreshed or deleted.
2. Send the API key as a bearer token in the Authorization header with each pplx-api request.
3. It currently support Mistral 7B, Llama 13B, Code Llama 34B, Llama 70B, and the API is conveniently OpenAI client-compatible for easy integration with existing applications.

For more information, visit the official Perplexity Labs API documentation and Quickstart Guide.

Filed Under: Technology News, Top News





Latest timeswonderful Deals

Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, timeswonderful may earn an affiliate commission. Learn about our Disclosure Policy.