DATABASES Signal 111
AI chip startup Positron raises $875M at $5B valuation for memory-first processor with up to 2.3TB on-chip memory
↗
Positron secured $875M in funding to develop Asimov, an AI inference processor with a memory-first architecture and up to 2.3TB of integrated memory.
This funding signals investor confidence in memory-centric AI hardware, which could reduce data movement bottlenecks in large-scale inference workloads. If successful, it may challenge traditional GPU-based architectures for memory-bound AI tasks.
Written by elseif from the cluster below · every claim links back to a sourceThe three things worth knowing
Positron’s Asimov processor targets AI inference with a memory-first design and up to 2.3TB of on-chip memory.
The $875M funding round values the startup at $5B, reflecting strong investor interest in alternative AI hardware.
Memory-first architectures could address latency and bandwidth constraints in large AI models
THE CLUSTER