AI News and Tutorials
Claude Opus 5 is officially launched: the performance has been greatly upgraded, achieving Fable 5 level performance at half the cost
I really didn’t expect that Claude Opus 5 would come out so quickly. The model update speed is really not...
Huang Renxun issued his first tweet on X: 25 technology companies signed an open letter calling on the United States to support open models
NVIDIA CEO Jensen Huang sent his first tweet on the X platform on July 25. The content was a letter...
Meta’s first big customer for renting idle computing power appears: Anthropic spends tens of billions to rent computing power from Meta
As we all know, Claud Code often suffers from instability or intelligence degradation due to excessive usage and insufficient computing...
AMD Advancing AI 2026: Helios cabinet-level AI system officially enters mass production, Anthropic, OpenAI, Meta team up to create megawatt-level computing power deployment
AMD officially released its first cabinet-level AI at the Advancing AI 2026 annual conference on July 24
Microsoft is testing a new feature that lets Copilot key remapping work for nothing
Windows 11 recently added a symbolic new setting to the Insider beta version: users can now reassign the Copilot-specific keys...
Doubling your efficiency while speaking! The desktop version of ChatGPT has officially added the ChatGPT Voice function, allowing AI to help you do things by just “speaking”
OpenAI brought a pretty good update earlier. ChatGPT Desktop App added the new function "ChatGPT Voice". When asking AI to...
OpenFPM lets CUDA run on Apple Metal GPU: 3D SPH runs ten times faster, and it takes 6 hours to get through the bottom layer with GPT-5.6 Sol
NVIDIA’s highest ecological moat CUDA may no longer be completely unshakable, OpenFPM developer Abhinav
Microsoft open source Mage-Flow image model: only 4B parameters can generate 1024×1024 images in one second, and the benchmark test is tied with FLUX.2 32B
Microsoft Asia publicly released Mage-Flow on July 22 at Hugging Face
Unsloth officially supports AMD GPU: AI training speed is increased by 2 times, VRAM usage is reduced by 70%, and LLM can be fine-tuned with 3 GB
In the field of AI development, NVIDIA’s CUDA ecosystem has long dominated the field, and most of the training and...
Microsoft VibeVoice-ASR: an open source speech recognition model that transcribes 60 minutes of audio files in one go and provides three-in-one structured output
Microsoft Research recently open sourced VibeVoice, a 7 billion-parameter speech AI model that can transcribe up to 60 words in...