Search AI Products and News

The Arc Prize Foundation, responsible for maintaining and managing the competition, last week revised its cost estimate for OpenAI's upcoming o3 inference AI model with a staggering adjustment—from an initial estimate of $3,000 per ARC-AGI task to $30,000. This price correction reveals that the operational costs of today's most complex AI models may be ten times higher than previously anticipated. While OpenAI has yet to announce an official pricing strategy for o3, or even officially release the model, the Arc Prize...

Following the release of the official Tencent HunYuan T1 last week, many users have been eager to know when this new version would be available on Yuanbao. In the latest news, Tencent officially announced that the official version of HunYuan T1 and the latest version of DeepSeek V3 are now available, bringing users a brand-new experience. HunYuan T1 is Tencent's self-developed deep thinking model, and compared to the previous T1 Preview version, it has undergone comprehensive upgrades. The new version not only improves speed and performance but also significantly enhances overall effectiveness, achieving second-level response times, and...

Former Intel CEO Pat Gelsinger recently criticized Nvidia's pricing strategy for its AI GPUs on the Acquired podcast during Nvidia's 2025 GPU Technology Conference, arguing that the high cost makes them unsuitable for large-scale AI inference tasks. Gelsinger highlighted that inference is crucial for deploying AI models and that the industry should focus more on inference, a domain where Nvidia's technology falls short in terms of cost-effectiveness. Image source omitted.

Reka AI, founded by a dozen former Google DeepMind scientists, has unveiled its first open-source model: Reka Flash 3. This 21-billion parameter inference model has garnered significant attention. Despite its relatively smaller parameter count, Reka Flash 3 is a general-purpose reasoning model trained from scratch. It underwent supervised fine-tuning on synthetic and public datasets and further refinement through model-based techniques.

OpenAI today announced the release of its latest high-performance inference model, O1-Pro. OpenAI states that the model offers significant improvements in performance and capabilities and is currently available to a select group of developers. O1-Pro is designed for developers building applications with 1-5 layers, supporting features like visual input, function calling, and structured outputs. It seamlessly integrates with OpenAI's response and batch processing APIs. However, this enhanced performance comes at a cost...

Tencent's Hunyuan officially announced the upcoming release of its new inference model, T1, at 11 PM Beijing time on March 21st. This marks another significant technological iteration and product upgrade in Tencent's AI large model field. Simultaneously, Tencent announced that its Hunyuan large model has successfully entered the top 15 globally on the renowned Chatbot Arena ranking. Chatbot Arena is a widely recognized anonymous evaluation platform for large models, where users anonymously interact with multiple models.

OpenAI has announced the launch of o1-pro, a new inference AI model within its developer API. OpenAI states that o1-pro leverages significantly more compute resources than its existing o1 model, aiming for consistently superior responses. However, this high-performance model is currently only available to a select group of developers – specifically, those who have spent at least $5 on OpenAI API services. Notably, o1-pro is significantly more expensive, costing ten times as much as the o1 model.

Nvidia has introduced Dynamo, a new software designed to significantly accelerate the performance of its DeepSeek AI technology, promising a 30-fold speed increase.

Recently, Professor Zhai Jidong's team from the Institute of High Performance Computing at Tsinghua University, in collaboration with Tsinghua-affiliated innovative enterprise, Qingcheng Extreme Intelligence, announced the open-sourcing of Chitu, a groundbreaking large model inference engine. This innovative technology marks another significant breakthrough for China in the AI field, particularly in inference engine development. A core highlight of the Chitu engine is its ability to natively run FP8 precision models on non-Nvidia Hopper architecture GPUs and various domestic chips.

New York-based AI team Nous Research, known for its personalized, unrestricted language models, has launched a new model inference API. This marks a significant step for the unconventional AI group, making its advanced language models accessible to a wider range of developers and researchers through a programmatic interface. The unrestricted nature of these models has always set Nous Research apart, challenging the likes of OpenAI and Anthropic...

Frustrated by the slow processing speed of large language models on long documents? Researchers from Tsinghua University have unveiled a groundbreaking technology – the APB parallel inference framework – that dramatically accelerates processing. Benchmark tests show this technology achieves a 10x speed improvement over Flash Attention when handling ultra-long texts. With the rise of models like ChatGPT, AI's ability to process vast amounts of text (hundreds of thousands of words) has increased significantly. However, this often comes at the cost of processing speed...

Silicon-Based Flow's official WeChat account announced that, effective immediately, the DeepSeek-R1 & V3 APIs on the SiliconCloud platform now support batch inference (BatchInference). Users can send requests to SiliconCloud via the batch API, unaffected by real-time inference rate limits, with task completion expected within 24 hours. Compared to real-time inference, DeepSeek-V3 batch inference pricing is reduced by 50%. From March 11th to March 18th, DeepSeek-R1 will enjoy a 75% price reduction.

SiliconCloud's platform now officially supports batch inference for DeepSeek-R1 and V3 APIs. Users can submit requests via batch APIs to process large datasets within 24 hours, overcoming the limitations of real-time inference rates. A key highlight is the significant price reduction; DeepSeek-V3 batch inference is substantially cheaper than real-time inference.

Cerebras Systems recently announced the construction of six new data centers across North America and Europe to bolster its artificial intelligence (AI) inference capabilities. This expansion significantly increases the company's computing power, supporting the growth of various AI applications. The plan allocates 85% of computing capacity to the US, with three facilities already operational in Santa Clara and Stockton, California, and Dallas, Texas. Upcoming centers will be located in Minneapolis (expected Q2 2025), Oklahoma City, and Montreal.