Exclusive: Positron raises $230M Series B to take on Nvidia’s AI chips

Positron is an AI chip startup that has raised $230M in Series B funding from investors including the Qatar Investment Authority. The company aims to develop and commercialize high-performance AI accelerators to compete with Nvidia's dominant position in the AI hardware market. With Qatar's strategic interest in building AI infrastructure and growing global demand for alternative AI chips, Positron is positioned to capture market share in the rapidly expanding AI compute sector. The company plans to design chips optimized for large language models, deep learning, and other AI workloads while offering better performance-per-watt and cost efficiency than current market leaders.

Positron positions itself as a specialized AI chip provider offering superior price-to-performance ratios compared to Nvidia's general-purpose GPUs. The company focuses on designing chips specifically optimized for AI/ML workloads rather than gaming or graphics applications. Key positioning points include: energy-efficient architectures for sustainable AI computing, custom silicon designed for next-generation AI models, and strategic partnerships with Middle Eastern markets through Qatar's AI infrastructure initiatives. Positron targets enterprise customers seeking to reduce dependence on single-vendor solutions while achieving better total cost of ownership.

• Specialized tensor processing units (TPUs) optimized for AI inference and training
• Energy-efficient architecture with superior performance-per-watt metrics
• Custom hardware support for large language models and transformer architectures
• Scalable chiplet design enabling flexible configurations for different workloads
• Advanced memory bandwidth and on-chip interconnect technology
• Compatible software stack with popular AI frameworks (PyTorch, TensorFlow)

Nvidia represents the primary competitor with its established CUDA ecosystem, comprehensive software stack, and market dominance. Other competitors include AMD (MI300 series), Intel (Gaudi accelerators), and specialized startups like Cerebras, SambaNova, and Graphcore. Cloud providers developing custom chips (Google TPU, AWS Trainium/Inferentia) also compete in certain segments. Positron's differentiation strategy focuses on: 1) Specialization in specific AI workloads rather than general-purpose GPU computing, 2) Strategic partnerships with Middle Eastern markets, 3) Energy efficiency advantages for sustainable AI deployments, and 4) Competitive pricing to undercut established players while maintaining performance parity.

Positron operates on a hardware-as-a-product model, selling AI accelerator chips and associated systems to enterprise customers, cloud providers, and research institutions. Revenue streams include: 1) Direct chip sales to hyperscale data centers and cloud providers, 2) Integrated system sales for enterprise AI deployments, 3) Licensing of chip designs and IP to strategic partners, and 4) Custom chip development services for specific use cases. The company may also explore a chip-as-a-service model where customers pay based on actual compute consumption rather than upfront hardware purchases. With $230M in Series B funding, Positron has sufficient runway for R&D, manufacturing partnerships (likely with TSMC or Samsung), and market expansion over the next 2-3 years.