Is AI Large Model Really Leading a “Capital Feast” in China? Yes and No

The actual popularity of large AI models in the primary market is far lower than that on social media. Despite frequent reports on social media about large AI models leading an investment boom in China, the actual situation is not the case. In the primary market, investors are cautious about the AI track and have limited willingness to invest in large models. It is very difficult for startups to obtain financing in the AI track, and the number of unicorns is also small. Internet giants make internal investments in large AI models, but it is hard to call it real "investment". The hype of large AI models in the capital circle is mainly limited to the secondary market of A-shares. The AI track faces problems within the industry and problems with venture capital supply, and neither of these two problems will be solved in the short term. Therefore, the claim that large AI models are leading a "capital feast" in China is seriously overestimated.

DeepMind Co-founder: AI Capabilities Are Overstated, Long-term Potential Is Underrecognized

DeepMind co-founder Demis Hassabis pointed out in a podcast that current AI capabilities are overstated, but its long-term potential, particularly in terms of General Artificial Intelligence (AGI), is still underestimated. The public's focus on AI has created a double-edged sword effect, showcasing its positive role in future everyday life while also generating a lot of noise. Hassabis appreciates AI models' ability to infer information about the real world through language learning, but he expressed concern over the public's overly high expectations of AI's short-term capabilities.

Midjourney Web Version Update: Launching a New Image Editor to Tackle the FLUX.1 Challenge

Midjourney has significantly upgraded its web-based image editing features, introducing a new integrated interface that combines local repainting, image aspect ratio adjustment, and prompt editing. This enhancement improves operational efficiency and user experience, responding to user demands and recent competitive pressures, particularly from the FLUX.1 tool. By optimizing its core image generation capabilities and fostering functional innovati....

Making Large Models Understand You Better: Tencent and Shanghai Jiao Tong University Join Forces to Decode the Secrets of Instruction Tuning

Tencent Youtu Lab and Shanghai Jiao Tong University released a comprehensive review focusing on dataset evaluation and selection for instruction tuning to enhance large model performance. The study aims to help large models grasp the essence of natural language processing by analyzing data quality, diversity, and importance to guide the construction of efficient and critical datasets. It highlights that data quality directly impacts tuning effect....

Geekbench Launches AI Performance Testing Tool: A New Standard for Evaluating Device AI Capabilities

Geekbench AI is a brand new cross-platform tool designed to evaluate device performance under AI-intensive workloads. It tests the CPU, GPU, and NPU to determine the device's ability to handle machine learning applications. It supports various AI frameworks such as ONNX, CoreML, TensorFlow Lite, and OpenVINO, and evaluates performance from both speed and accuracy dimensions, providing scores in full precision, half precision, and quantized precision. Additionally, Geekbench AI's testing results...

The New Geekbench AI Benchmark Tests the Performance of CPU, GPU, and NPU

Geekbench AI is a new cross-platform tool designed to evaluate the performance of devices under compute-intensive workloads, with a particular focus on machine learning tasks. It provides full precision, half precision, and quantized scores by measuring the performance of CPUs, GPUs, and NPUs, and includes accuracy assessments. It supports various machine learning frameworks, including ONNX, CoreML, TensorFlow Lite, and OpenVINO. This tool is available on Windows, macOS, Linux, Android, and iOS.