Tongxin Intelligence, incubated by the Beijing Institute for General Artificial Intelligence and initiated by Zhu Songchun, has closed seed and angel rounds totaling hundreds of millions of RMB within six months of founding, with proceeds going to pre-tapeout architecture design. The company is building a novel inference chip for AGI algorithms that stays compatible with mainstream models while adding dedicated hardware support for logical, causal and value reasoning, aiming to define next-generation compute before the algorithms settle.
Inference Chip
推理芯片
Dedicated chip designed for the model inference stage, emphasizing low latency and high energy efficiency.
Curated Intelligence
AI cloud provider General Compute secured a $100 million Series A round amid broader neocloud market skepticism. The company differentiates itself by building infrastructure on specialized inference chips from Nvidia rivals like SambaNova and Positron, focusing on unit economics and profitability rather than just raw speed.
Hugging Face Text Generation Inference (TGI) has announced support for AWS Inferentia2 chips. This update expands TGI's compatibility with cloud-specific inference hardware, though the source text lacks specific performance benchmarks, latency data, or cost optimization details.
AWS announced that its dedicated inference chip, Inferentia2, now supports accelerating Hugging Face Transformers models. This integration aims to enhance LLM inference speed and cost-efficiency through hardware optimization, offering developers a more economical cloud deployment solution.