Chinese AI darling DeepSeek is back with a new open weights large language model that promises performance to rival the best proprietary American LLMs. Perhaps more importantly, it claims to dramatically reduce inference costs and it extends support for Huawei's Ascend family of AI accelerators.

Google LLC introduced two new custom silicon chips for artificial intelligence today at Google Cloud Next 2026, unveiling two distinct Tensor Processor Unit architectures built for training and inference: the eighth-generation TPU 8t and TPU 8i.

Ahead of COMPUTEX 2026, Skymizer Taiwan Inc., a pioneer in AI inference solutions, today previewed a major advancement in on-premise AI deployment with its HTX301 inference chip, which integrates HyperThought™ — a software/hardware co-design platform first introduced at COMPUTEX 2025.

QVAC SDK and Fabric give people and companies the ability to execute inference and fine-tune powerful models on their own terms, on their own hardware, with full control of their data.” Paolo Ardoino,

WEST PALM BEACH, Fla.--(BUSINESS WIRE)--Vultr, the world’s largest privately-held cloud computing platform, today announced the launch of Vultr Cloud Inference. This new serverless platform revolutionizes AI scalability and reach by offering global AI ...

As demand for open-source AI infrastructure grows, Novita AI is establishing itself as the inference provider for developers and engineering teams that need fast and affordable inference for production AI.

AI reasoning does not necessarily require spending huge amounts on frontier models. Instead, smaller models can yield stronger performance on complex tasks while keeping per-query inference costs mana

Forbes contributors publish independent expert analyses and insights. I write about the economics of AI. When OpenAI’s ChatGPT first exploded onto the scene in late 2022, it sparked a global obsession with model training. Companies raced to build ever ...