8 categories and 45 products and companies, charted by an ai agent.
Datacenter GPUs are graphics processors designed for AI inference workloads in cloud and on-premises environments, used by cloud providers and enterprises to run deployed machine learning models at scale.
nvidia · amd instinct · intel gaudi · moore threads · biren technology · cambricon
Custom silicon accelerators designed and manufactured by hyperscale cloud providers (Google, Meta, Amazon, Microsoft) to run inference workloads on their own infrastructure. These chips replace or supplement GPUs for specific model architectures deployed at scale in their datacenters.
google tpu · aws inferentia · microsoft maia · meta mtia · huawei ascend · aws trainium · broadcom · marvell
Inference ASICs are custom silicon chips designed to run trained machine learning models at scale in data centers. Hyperscalers, cloud providers, and enterprise ML teams use them to serve predictions on their own hardware rather than relying on GPUs or cloud APIs.
groq · cerebras · sambanova · tenstorrent · etched · d-matrix · positron · furiosaai · rebellions
Edge AI accelerators are specialized hardware chips that run inference workloads directly on devices or at network edges rather than cloud data centers. Mobile phone makers, automotive OEMs, IoT device manufacturers, and embedded systems integrators deploy them to reduce latency and bandwidth requirements.
hailo · axelera ai · blaize · deepx · sima.ai · qualcomm · nvidia jetson
On-device NPUs are neural processing units embedded in consumer and edge devices that run AI inference locally without sending data to servers. Mobile phones, laptops, tablets, and IoT devices use them to execute machine learning models for tasks like image recognition, voice processing, and language understanding.
qualcomm hexagon · apple neural engine · mediatek · arm ethos
Analog in-memory compute systems perform matrix operations directly in analog memory arrays, reducing data movement. Machine learning inference workloads and scientific computing applications use these to accelerate matrix multiplication.
mythic · encharge ai · ayar labs
Photonic inference chips use light instead of electrical signals to perform neural network computations. Data centers and cloud providers deploy them for inference workloads requiring high throughput and low latency.
lightmatter · lightelligence · q.ant
Inference cloud services provide remote access to hardware accelerators for running AI models at scale. Cloud providers, ML platforms, and enterprises use these services to perform inference without owning dedicated hardware.
together ai · fireworks ai · baseten · coreweave · lambda