CMP Gadgets

Tag: chatgpt

  • OpenAI Unveils Ultrafast Mode for GPT-5.6 Sol

    OpenAI Launches Ultrafast Mode for GPT-5.6 Sol

    OpenAI introduced a new Ultrafast mode on Tuesday. The company released it as an early preview. This tier aims to cut the time GPT-5.6 Sol needs to finish tasks. It generates up to 750 output tokens per second.

    The ChatGPT maker claims Ultrafast mode runs GPT-5.6 Sol far quicker than standard processing. OpenAI is currently working with select customers. These partners are evaluating the new tier’s performance.

    In a blog post, OpenAI announced the early preview of Ultrafast. The new API service tier can run GPT-5.6 Sol up to 14 times faster than regular processing. Cerebras powers the GPT-5.6 Sol model in Ultrafast mode. The setup generates as many as 750 output tokens each second.

    This speed lets businesses create more responsive products. Companies can also make faster decisions and bring strong AI into demanding workflows. GPT-5.6 Sol on Ultrafast mode is available today in a limited preview. Only a select group of customers can access it now. OpenAI plans to expand access as capacity grows.

    The company highlighted several potential uses. These include real-time incident response, market signal analysis, transaction assessment, customer support, and live research. OpenAI is working with an initial group of customers. The goal is to learn where this speed creates the biggest difference. Those insights will help shape future products.

    OpenAI is testing the mode across coding, commerce, financial research, and customer support applications. Early trials show where a major jump in speed adds value to workflows. The company will use these findings to guide wider deployment as capacity increases.

    Internal developers are also testing GPT-5.6 Sol in Ultrafast mode. Teams use it for incident response. The mode reduces the delay between spotting a signal, testing a hypothesis, and choosing the next step. Staff further apply it to research tasks. These involve searching knowledge sources, querying data, and organising or summarising information across connected tools.

  • Google Launches Gemini 1.5 Pro AI Models

    Google released newer versions of the Gemini 1.5 Pro artificial intelligence (AI) model on Tuesday. These arrive just months after the Mountain View-based tech giant released the last version of Gemini, which increased the context window to up to 2 million tokens. Dubbed Gemini-1.5-Pro-002 and Gemini-1.5-Flash-002, the company said these AI models not only offer higher output and lower costs but will also provide a higher rate limit to users. Additionally, the filter settings have also been updated to make the AI models adhere to instructions more closely.

    The company detailed the Gemini-1.5-Pro-002 and the Gemini-1.5-Flash-002 AI models. These models are currently available as experimental model releases and are built on top of Gemini 1.5 Pro which was first released at Google I/O in May. Currently, these are available to developers and enterprise customers of the company. Developers can access it for free from Google AI Studio and the Gemini API. Enterprises can access it via Vertex AI.

    The latest Gemini 1.5 Pro and Flash models also outperformed the previous generation’s Gemini model, as per internal testing conducted by Google. The company claimed that the new models noted a seven percent increase in the Massive Multitask Language Understanding Pro (MMLU-Pro) benchmark. Further, the AI models are said to offer roughly 20 percent improvement on MATH and HiddenMath benchmarks compared to Gemini 1.5 Pro.

  • Gemini AI: News

    Gemini AI for Android is reportedly getting a new feature allowing users to edit the images generated by the chatbot. A report found evidence for the feature within the Google beta app for Android. Interestingly, the feature is said to allow users to edit images using both text prompts and by highlighting a specific part of the image. While the feature was found in the beta app of Google, it has not been made visible and beta testers might not be able to use it.

    Editing the images generated by an artificial intelligence (AI) chatbot is a difficult proposition at present. In most cases, users just have the option to regenerate the image with improved prompts. However, there is no guarantee the newly generated images will be similar to the ones previously created by the chatbot. Last week, Meta announced image editing capabilities for its chatbot.

  • Gemini 1.5 Flash AI Model

    Gemini 1.5 Flash artificial intelligence model will now power the Gemini chatbot, Google announced on Thursday. With this update, the free tier of the AI chatbot on the web and the app will now be powered by the new model. Earlier, the chatbot was powered by the Gemini 1.0 Pro large language model (LLM). The tech giant claims that the new model is faster and more efficient compared to its predecessor. Notably, the update arrived just days after Meta added new capabilities to Meta AI.

    The tech giant announced the rollout of the new AI model in Gemini’s release notes. As per the post, the Gemini 1.5 Flash increases the context window of the chatbot from the previous 8,000 tokens to 32,000 tokens. With the upgraded context window, Gemini should be able to handle more complex queries and add more context to its generated responses.

    First announced by the company during its Google I/O event in May, the Google 1.5 Flash is a lightweight model which was distilled from the Gemini 1.5 Pro. At the time, the company highlighted that the Flash model prioritizes speed and reduces the input and output processing costs.

  • Claude artificial intelligence (AI)

    Anthropic launched an Android app for its Claude artificial intelligence (AI) chatbot on Tuesday. The Android app comes a couple of months after the AI firm released an iOS app for Claude. Users can find the Claude AI app free to download on the Play Store. The interface works similarly to the website, and even free users can take advantage of all the basic features. The chatbot is powered by Claude 3.5 Sonnet by default, which was released by the company last month.

    The AI firm announced the launch of the Android app via a post on X (formerly known as Twitter). Unlike the iOS app release, Anthropic did not specify if this was a global release or would only be available in select regions. Gadgets 360 was able to access the app through the Play Store. Notably, the Claude AI app is 4.12MB in size.

    Claude’s Android app is free to download and use. However, users will have to sign up using an email and register their account with their phone number. Free users will be able to access all the basic features of the text-based AI model. Since Claude 3.5 Sonnet powers the chatbot now, users will also be able to access its computer vision capabilities. To use the feature, users will need to give the app access to the camera and the gallery. Once done, they can click any picture and ask the AI for information about it.

  • Nothing Phone 2a

    Nothing Phone 2a recently received the Nothing OS 2.5.5 update with new camera features, general improvements, bug fixes, and Google’s April security patch. Now, the Nothing OS 2.5.5a update is rolling out for Phone 2a users worldwide with ChatGPT integration and more new features.

    In a post on X, Nothing confirmed that the Nothing OS 2.5.5a update is rolling out for Phone 2a users. One of the key elements in this changelog is the ChatGPT integration. To be able to access all features, they have to ensure that the latest version of the ChatGPT application is downloaded on their Phone 2a handsets from the Play Store.

    After installing the Nothing OS 2.5.5a update, Phone 2a users will be able to access ChatGPT via their Nothing Ear or Nothing Ear A true wireless earphones. The Nothing X application has an option to customize a gesture to start a direct voice conversation with ChatGPT.

  • Google’s AI Gemini: New Feature

    Google added its native artificial intelligence (AI) chatbot to the Android operating system as an on-device voice assistant earlier this year. This feature gave users the option to replace Google Assistant with Gemini as the default assistant for the smartphone. However, there were certain drawbacks to using the AI chatbot as it cannot set alarms, add reminders or launch apps like its counterpart. A new report has now found that the tech giant might soon add the capability for it to play music with a simple voice command.

    The new feature was spotted within the settings options of the Gemini app for Android by a PiunikaWeb report (via AssembleDebug). Based on screenshots shared, a new option was seen within Gemini Settings. The second option from the bottom now has a Music option which features the description “Select preferred services to play music”. Another screenshot shows that within the option, another title says, “Choose your default media provider”.

  • Meta Llama 3

    Meta introduced the next generation of its AI models, Llama 3 8B and 70B, on Thursday. Shortened for Large Language Model Meta AI, Llama 3 comes with improved capabilities over its predecessor. The company also adopted new training methods to optimize the efficiency of the models. Interestingly, with Llama 2, the largest model was 70B, but this time the company said its large models will contain more than 400 billion parameters. Notably, a report last week revealed that Meta will unveil its smaller AI models in April and its larger models later in the summer. Those interested in trying out the new AI models are in luck as Meta is taking a community-first approach with the Llama 3. The new foundation models will be open source just like previous models. Meta stated in its blog post, “Llama 3 models will soon be available on AWS, Databricks, Google Cloud, Hugging Face, Kaggle, IBM WatsonX, Microsoft Azure, NVIDIA NIM, and Snowflake, and with support from hardware platforms offered by AMD, AWS, Dell, Intel, NVIDIA, and Qualcomm.” The list includes all major cloud, hosting, and hardware platforms, which should make it easier for enthusiasts to get their hands on the AI models. Further, Meta has also integrated Llama 3 with its own Meta AI that can be accessed via Facebook Messenger, Instagram, and WhatsApp in supported countries.