Phison Collaborates with Intel to Bring Larger Local AI Workloads to Intel AI PC Platforms
View original at finance.yahoo.comPhison Collaborates with Intel to Bring Larger Local AI Workloads to Intel AI PC Platforms Phison and Intel collaborate to bring larger local AI workloads to Intel AI PC Platforms Phison's Pascari aiDAPTIV™ removes local memory constraints on client PCs, enabling larger MoE AI models and agentic AI applications TAIPEI,…
Lo que extrajimos de esta fuente
The claims Via News extracted from this document. We point to the source; we don't replace it.
AI PCs are evolving into platforms for more sophisticated local AI workloads, and aiDAPTIV helps expand available memory for AI workloads on Intel AI PC platforms, allowing OEMs, developers and end users to run more capable AI applications locally while maintaining privacy and infrastructure efficiency.
60% confidencePhison and Intel are working to support ISV evaluations, technical demonstrations and optimized workloads for public performance claims.
60% confidenceIn Phison testing, aiDAPTIV enabled a 26B-parameter model to run on a system with 16GB of DRAM, compared with 32GB of DRAM required without aiDAPTIV in the same test environment.
60% confidenceaiDAPTIV addresses memory constraints by extending effective AI working memory across system DRAM and high-performance, extreme-endurance NAND flash, reducing DRAM required for certain local AI workloads and supporting runtime features such as KV cache reuse.
60% confidence
Citado en estos reportajes de Via News
- Cuando el chip se convierte en bono: la fiebre de la IA reescribe cómo se financian los semiconductores →
- El auge de los chips de IA financiariza el sector: Nvidia propone titulizar GPU por 500.000 millones mientras Micron dispara su inversión →
- El superciclo de los chips de IA sigue firmando acuerdos multimillonarios mientras Wall Street castiga al sector con una corrección del 25% →
- La fiebre de los chips de IA se enfría en Wall Street: qué significa para los inversores hispanohablantes cuando ni un solo protagonista del expediente es de habla hispana →
- Nvidia propone titulizar chips por 500.000 millones de dólares mientras Washington financia a GlobalFoundries: el boom de la IA obliga a reinventar cómo se paga el hardware →
