Tech

Baidu released the Ernie 4.5 series of artificial intelligence (AI) models in open-source on Monday. The Chinese tech giant had previously said that it would make its proprietary large language models (LLMs) available to the open community on July 31. It has now released 10 different variants of the series, with each of the models built on Mixture-of-Experts (MoE) architecture. Alongside the models, the company has also released multi-hardware development toolkits for Ernie 4.5 in open source.

Baidu Releases 10 Variants of Ernie 4.5 AI Models in Open Source

In a post on X (formerly known as Twitter), the Chinese tech giant announced the release of the 10 open-source Ernie 4.5 AI models. Four of them are multimodal vision-language models, eight are MoE models, and two are thinking or reasoning models. Additionally, the list also includes five post-trained models, while others are pre-trained. These models can now be downloaded from either the company’s Hugging Face listing or from its GitHub listing.

In a blog post, Baidu said that the MoE models feature a total of 47 billion parameters, with three billion of them being active at a time. The largest models among the 10 variant features 424 billion parameters. All of them are trained using the PaddlePaddle deep learning framework.

Based on internal testing, the company claimed the Ernie-4.5-300B-A47B-Base model surpasses DeepSeek-V3-671B-A37B-Base on 22 out of 28 benchmarks. Similarly, it claimed that the Ernie-4.5-21B-A3B-Base outperforms Qwen3-30B-A3B-Base on multiple mathematics and reasoning benchmarks despite having 30 percent fewer parameters.

Baidu also revealed its training methods on the model pages. The company used a heterogeneous MoE structure in the pre-training process and scaled the models using techniques such as intra-node expert parallelism, memory-efficient pipeline scheduling, FP8 mixed-precision training, and a fine-grained recomputation method.

Apart from the models, Baidu has also made ErnieKit available to the open community. It is a development toolkit for the Ernie 4.5 series models. With this, developers can perform pre-training, supervised fine-tuning (SFT), Low-Rank Adaptation (LoRA), and other customisation techniques. Notably, all the models are available under the permissive Apache 2.0 licence, which allows for both academic and commercial usage.

Apple must face the US Department of Justice’s lawsuit accusing the iPhone maker of unlawfully dominating the US smartphone market, a judge ruled on Monday.

US District Judge Julien Neals in Newark, New Jersey, denied Apple’s motion to dismiss the lawsuit accusing the company of using restrictions on third-party app and device developers to keep users from switching to competitors and unlawfully dominate the market.

The decision allows the case to go forward in what could be a years-long fight for Apple against enforcers’ attempt to lower what they say are barriers to competition with Apple’s iPhone.

An Apple spokesperson said the company believes the lawsuit is wrong on the facts and the law, and will continue to vigorously fight it in court.

A spokesperson for the DOJ declined to comment.

Sales of the world’s most popular smartphone totaled $201 billion (roughly Rs. 17,19,518 crore) in 2024. Apple introduced a new budget model iPhone in February with enhanced features priced at $170 more than its predecessor.

The lawsuit filed in March 2024 focuses on Apple’s restrictions and fees on app developers, and technical roadblocks to third-party devices and services — such as smart watches, digital wallets and messaging services — that would compete with its own.

DOJ, along with several states and Washington, DC, says the practices destroy competition and Apple should be blocked from continuing them.

Apple had argued that its limitations on third-party developers’ access to its technology were reasonable, and that forcing it to share technology with competitors would chill innovation.

The case is one of a series of U.S. antitrust cases against Big Tech companies brought during the Biden and first Trump administrations.

Facebook parent Meta Platforms and Amazon.com are facing lawsuits by antitrust enforcers alleging they illegally maintain monopolies, and Alphabet’s is facing two such lawsuits.

© Thomson Reuters 2025

OpenAI said it has no active plans to use Google’s in-house chip to power its products, two days after Reuters and other news outlets reported on the AI lab’s move to turn to its competitor’s artificial intelligence chips to meet growing demand.

A spokesperson for OpenAI said on Sunday that while the AI lab is in early testing with some of Google’s tensor processing units (TPUs), it has no plans to deploy them at scale right now.

Google declined to comment.

While it is common for AI labs to test out different chips, using new hardware at scale could take much longer and would require different architecture and software support. OpenAI is actively using Nvidia’s graphics processing units (GPUs), and AMD’s AI chips to power its growing demand. OpenAI is also developing its chip, an effort that is on track to meet the “tape-out” milestone this year, where the chip’s design is finalized and sent for manufacturing.

OpenAI has signed up for Google Cloud service to meet its growing needs for computing capacity, Reuters had exclusively reported earlier this month, marking a surprising collaboration between two prominent competitors in the AI sector. Most of the computing power used by OpenAI would be from GPU servers powered by the so-called neocloud company CoreWeave.

Google has been expanding the external availability of its in-house AI chips, or TPUs, which were historically reserved for internal use. That helped Google win customers, including Big Tech player Apple, as well as startups like Anthropic and Safe Superintelligence, two ChatGPT-maker competitors launched by former OpenAI leaders.

© Thomson Reuters 2025

Sony seems to have finally raised the price of the PS5 in India, weeks after the company hiked prices of its console in the UK, Europe and other markets. The company has only raised the price of the PS5 digital edition, just like it did in the UK and Europe in April. The price of the disc edition of the console remains the same.

PS5 Digital Edition Price Hiked

The price of the digital edition PS5 slim has gone up Rs. 5,000 — from Rs. 44,990 to Rs. 49,990 — as seen on Sony’s Shop at SC website (spotted by High Chaos Run). Curiously, the price of the PS5 digital edition Fortnite bundle remains at Rs. 44,990.

The price of the standard PS5, which comes with a disc drive, also stays the same, at Rs. 54,990. Prices of PS5 accessories like DualSense controllers and Pulse 3D wireless headsets have also not been hiked.

Third party retail listings of the PS5 digital edition on Amazon, Flipkart and other retail websites do not reflect the hiked pricing yet. Sony has not officially announced the price increase in India, but Gadgets 360 has reached out to PlayStation India for confirmation.

In April, Sony hiked PS5 prices in several key markets, including the UK, Europe, Middle East and Africa (EMEA), Australia, and New Zealand, citing “challenging economic environment”. In the UK and Europe, the company only hiked the price of the digital edition of the console.

At its quarterly earnings call in May, the PlayStation parent said it sold 18.5 million PS5 units in FY 2024, a decline from the 20.8 million units sold in the year before. Uncertain economic environment and widespread tariffs imposed by US President Donald Trump earlier this year led to several manufacturers raising prices of their products.

In May, Microsoft raised prices for Xbox Series S/X consoles and accessories and said some of its first-party games would cost $80 starting this holiday season. “We understand that these changes are challenging, and they were made with careful consideration given market conditions and the rising cost of development,” Xbox said at the time.

Xiaomi has been launching three models as a part of its flagship numbered series for the past few years. Last year, Xiaomi launched its Xiaomi 15 and Xiaomi 15 Pro models followed by its Xiaomi 15 Ultra in China. Out of these only the Xiaomi 15 and the Xiaomi 15 Ultra made it to India. A tipster now claims that we will see two Pro models this year along with the standard model and the Ultra model. This new model has been tagged as the Xiaomi 16 Pro Mini and we now have some details about this new compact flagship.

According to a post by Digital Chat Station on Weibo, Xiaomi is working on a new 16 Pro model that will be smaller in size compared to the regular Pro. While we do not have any dimensions about this mystery smartphone, the tipster mentions that it will have a 6.3-inch display versus the 6.8-inch display on the regular Pro model.

What’s new is the design of the rear camera module. Both Pro models are tipped to have large horizontal modules that will occupy one-third of the rear panel. This would be very similar to the camera design that Apple is reportedly going with for its iPhone 17 Pro models this year.

In terms of capability, the tipster claims that both the bigger Xiaomi 16 Pro and the compact Xiaomi 16 Pro Mini will have a periscope telephoto camera as well.

The tipster also shared an image which showcases the various glass treatments and camera layout options, which could make it to the new Xiaomi 16 Pro models. The camera module cutouts indicate the presence of three cameras with extra cutouts for the mic and other sensors. The designs vary by texture and clarity with some also having black ring or lens protectors for the three cameras embedded into the camera module.

While we now have an idea that Xiaomi is planning on launching an additional compact flagship. It still remains unclear as to why Xiaomi would add another smartphone to its numbered series with the same display dimensions as the standard numbered model. Adding a Pro Mini model makes sense only if Xiaomi plans to take away a few features or remove a camera from the standard model. The currently available Xiaomi 15 has three 50-megapixel rear-facing cameras. So, we could be looking at Xiaomi 16 with two rear-facing cameras, while the new Xiaomi 16 Pro Mini gets three.

Vi (Vodafone Idea) launched a new Vi Guarantee Programme for 2G handset customers on Tuesday. The new programme offers additional validity for prepaid customers. It is valid on Vi prepaid packs priced at Rs. 199 and Rs. 209. These recharge plans provide unlimited calls, 2GB of data, and 300 SMS. Notably, the Vi Guarantee programme for 4G and 5G users was announced last year. 

Vi Announced a New Vi Guarantee Offer

Vi’s new Vi Guarantee Programme offers two extra validity days with every unlimited voice recharge of Rs. 199 or more. Over 12 months, this adds up to 24 bonus days. Vi says the new initiative ensures a full 30 days of service per pack instead of the usual 28 days, reducing the need for frequent recharges.

The latest Vi Guarantee benefit is applicable for prepaid customers who are using 2G handsets and are on unlimited voice recharge packs of Rs. 199 and Rs. 209. The Rs. 199 plan comes with unlimited calls, 2GB data and 300 SMS. The plan has a validity of 28 days. Meanwhile, the Rs. 209 plan provides unlimited calls, 2GB data, 300 SMS, and caller tune offers.

Users in Assam, North East, Odisha, Himachal Pradesh, Jammu and Kashmir and Rajasthan get 3GB data and 300 SMS alongside unlimited calls with the Rs. 199 and Rs. 209 packs. The Rs. 209 pack offers caller tunes benefit as well. 

Vi customers can activate the Vi Guarantee benefit by dialling *999# or calling 1212. The extra data offered under the Vi Guarantee can be claimed through the Vi App.

The company had launched the Guarantee programme for 4G and 5G customers last year. This programme offers a total of 130GB extra data over one year on a complimentary basis.

Vi Guarantee Programme for 2G users launch follows closely on the heels of Vi’s recent 5G expansion across 23 Indian cities. The Vi 5G network is already live in major hubs such as Bengaluru, Delhi, Mumbai, Patna, and Chandigarh.

OnePlus Nord CE 5 is set to launch in India on July 8. Ahead of the launch, the company has revealed the key features of the phone, including its chipset, camera, battery, and charging details. It will debut alongside the OnePlus Nord 5 and the OnePlus Buds 4 TWS earphones. OnePlus has confirmed the battery life of the earbuds. Availability details for the upcoming devices have also been revealed, and some specifications of the Nord 5 and Buds 4 have already been confirmed.

OnePlus Nord 5 Series, OnePlus Buds 4 Availability

The OnePlus Nord 5 will go on sale in India on July 9 at 12pm IST, while the Nord CE 5 handset will be available for purchase starting July 12 at 12am IST, the company confirmed in a press release. The upcoming smartphones, alongside the OnePlus Buds 4, will be available via the OnePlus India website, Amazon and offline OnePlus stores.

OnePlus Nord CE 5 Key Features

The OnePlus Nord CE 5 will be equipped with a 4nm octa-core MediaTek Dimensity 8350 Apex chipset with an Mali-G615 GPU and LPDDR5X RAM. The smartphone will house a 7,100mAh battery with 80W SuperVOOC fast charging support. It is claimed to complete a full charge from one to 100 percent in just 59 minutes. A 10-minute charge is said to offer up to six hours of YouTube streaming. 

OnePlus revealed that the Nord CE 5 handset will support bypass charging. The phone will also feature Battery Health Magic, a charging system designed to “intelligently” manage charging to preserve battery health and extend its lifespan.

For optics, the OnePlus Nord CE 5 will be equipped with a 50-megapixel 1/1.95-inch Sony LYT-600 primary rear sensor with Optical Image Stabilisation (OIS) support. It will also support up to 4K video recording at 60fps. It will use the same RAW HDR and Real Tone technology from the OnePlus 13 series, the company added. The camera system will support Ultra HDR in Live Photos as well.

OnePlus Buds 4 Specifications

The OnePlus Buds 4 are claimed to offer a total battery life of up to 45 hours, together with the charging case. Meanwhile, the earbuds alone are said to last for up to 11 hours on a single charge. The TWS earphones will come with fast charging capabilities as well.

OnePlus Buds 4 will offer intuitive touch controls, including slide gestures on the earbud stems for volume control. They will support Steady Connect technology, which is said to ensure a stable Bluetooth connection even outdoors. The earphones will also support Dual-Device connectivity and Google Fast Pair along with an AI Translation feature.

Microsoft researchers unveiled a new artificial intelligence (AI) system on Monday that can diagnose patients more accurately than human doctors. Dubbed the Microsoft AI Diagnostic Orchestrator (MAI-DxO), it includes multiple AI models and a framework that allows it to go through patient symptoms and history to suggest relevant tests. Based on the results, it then suggests possible diagnoses. The Redmond-based tech giant highlighted that apart from the accuracy of the diagnosis, the system is also trained to be cost-effective in terms of tests conducted.

Microsoft Develops Benchmark to Test MAI-DxO’s Performance

In a post on X (formerly known as Twitter), Mustafa Suleyman, the CEO of Microsoft AI, posted about the MAI-DxO system. Calling it a “big step towards medical superintelligence,” he said the AI system can solve some of the world’s toughest medical cases with higher accuracy and lower costs compared to traditional diagnostic measures.

MAI-DxO simulates a virtual panel of physicians with diverse diagnostic approaches who collaborate to solve medical cases, the company said in a blog post. The Orchestrator includes a multi-agentic system where one provides a hypothesis, one picks the tests, two others provide checklists and stewardship, and the last challenges the hypothesis.

MAI-DxO workflow
Photo Credit: Microsoft

Once a hypothesis passes this panel, the AI system can either ask a question, request tests, or provide the diagnosis if it feels it has enough information. In case it recommends a test, it performs a cost analysis to ensure that the overall cost remains reasonable. Interestingly, the system is model agnostic, meaning it can perform with any third-party AI models.

Microsoft claims that the system boosts the diagnostic performance of every AI model that was tested. However, OpenAI’s o3 fared the best by correctly solving 85.5 percent of the New England Journal of Medicine (NEJM) benchmark cases. The company said that the same cases were also given to 21 practising physicians from the US and UK, and all of them had between five to 20 years of clinical experience. The human doctors had an accuracy of 20 percent.

MAI-DxO can be configured to operate within defined cost constraints, the company said. Once an input budget has been added, the system explores cost-to-value trade-offs while making diagnostic decisions. This helps in the AI system only ordering the necessary tests, instead of every possible test to rule out all causes of the symptoms.

To assess the AI system, Microsoft also developed a new benchmark dubbed the Sequential Diagnosis Benchmark (SD Bench). Unlike typical medical benchmark tests that ask multiple-choice questions, this test assesses AI systems’ ability to iteratively ask the right questions and order the right tests. Then it evaluates the answers by comparing them to the outcome published in the NEJM.

Notably, the MAI-DxO is not yet approved for clinical use, and is meant as initial research into developing AI capability in diagnostic operations. Microsoft said that its AI system can only be approved for clinical usage after rigorous safety testing, clinical validation, and regulatory reviews.

Honor is expected to unveil its Magic 8 series later this year. The handsets in the purported lineup could be equipped with the unannounced Snapdragon 8 Elite 2 SoC. The series may include four variants, including a Mini and a Max, with 6.3-inch and 6.9-inch screens, respectively. Notably, the current Honor Magic 7 and Magic 7 Pro variants use the Snapdragon 8 Elite chipset. They sport 6.78-inch and 6.8-inch displays, respectively.

Honor Magic 8 Series: All We Know

The Honor Magic 8 series is expected to boast 6.3-inch, 6.58-inch, and 6.7-inch displays, according to a Weibo post by Digital Chat Station (DCS). The 6.58-inch variant could be the base Honor Magic 8 phone, which is reportedly being tested with a mid-tier periscope camera, either the 64-megapixel OmniVision OV64B or the 50-megapixel Sony IMX882 sensor.

Honor Magic 8 Pro could sport a 6.7-inch display and feature a 200-megapixel periscope camera. Meanwhile, the smartphone with a 6.3-inch screen may be the Honor Magic 8 Mini. The smaller variant may not launch alongside the base and Pro models and could arrive at a later date.

An earlier leak by the same tipster suggested that the Honor Magic 8 series will likely include a Max variant, expected to feature a 6.9-inch display.

The Honor Magic 8 Pro has previously been tipped to be equipped with the yet to be announced Snapdragon 8 Elite 2 SoC. For optics, it may include a triple rear camera unit, including a 50-megapixel OV50Q sensor, a 50-megapixel sensor with an ultrawide-angle lens, and a 200-megapixel periscope telephoto sensor. The Magic 7 Pro carries a similar rear camera setup alongside a 50-megapixel selfie snapper.

Honor Magic 7 and Honor Magic 7 Pro with the Snapdragon 8 Elie SoC were launched in China with 6.78-inch and 6.8-inch displays, respectively. 

Redmi Turbo 4 Pro with a Snapdragon 8s Gen 4 SoC and 7,550mAh battery was launched in China earlier this year. Now, the Xiaomi sub-brand seems to be working on the Redmi Turbo 5 Pro as its successor. While the company is yet to confirm its existence, a new leak has surfaced out of China indicating that the Redmi Turbo 5 Pro will have a larger battery unit and a 6.8-inch display. It is expected to debut in global markets outside China as the Poco F8.

Tipster Digital Chat Station has claimed on Weibo that an upcoming Redmi smartphone will feature a 6.8-inch flat screen and a battery capacity exceeding 8,000mAh. The tipster doesn’t specifically mention the moniker of the phone, but the comments of the post point to a Redmi phone launching in 2026, which is likely to be the Redmi Turbo 5 Pro. The tipster might also be hinting at a successor to the Redmi Turbo 4 or Redmi K80 Ultra.

Notably, the Poco F7, which was unveiled in India and other global markets recently, shared a lot of its hardware with the Redmi Turbo 4 Pro. Therefore, it’s likely that the upcoming Poco F8 will be based on the Redmi Turbo 5 Pro. If that’s the case, it could also feature an 8,000mAh battery and a 6.8-inch flat display, as suggested by the tipster. 

The Redmi Turbo 4 Pro was launched in China in April this year. The Xiaomi sub-brand introduced the Poco F7 5G in India last month with a price tag of Rs. 31,999 for the 12GB RAM + 256GB storage model.

Redmi Turbo 4 Pro, Poco F7 Specifications

Both Redmi Turbo 4 Pro and Poco F7 5G feature the Snapdragon 8s Gen 4 chipset, 6.83-inch OLED displays with 1.5K resolution and 120Hz refresh rate, 50-megapixel dual rear camera units, and 20-megapixel selfie shooters. They pack a 7,550mAh battery with 90W wired fast charging and 22.5W wired reverse charging support. The phones are said to meet the IP66+IP68+IP69 ratings for dust and water resistance. They feature an in-display optical fingerprint sensor for authentication. The main difference between the two phones is the design.