EVAS Intelligence: AI Compute Chips and Hardware Solutions
EVAS provides AI acceleration hardware for enterprise and automotive applications, including dedicated computing for smart cockpits and driving systems.
At a glance
- Category
- Software Development
- Best for
- Automotive manufacturers, Enterprise AI developers, Hardware integration teams, Software companies building large-model applications
- Pricing
- Pricing was not clearly available from the provided evidence. Buyers should confirm current pricing on the vendor website.
- Key use cases
- Automotive Intelligent Cockpits, Intelligent Driving Systems, Enterprise AI Inference, Centralized Vehicle Computing
- Integrations
- PyTorch, TensorFlow, PCIe 5.0
- Official website
- evas.ai
EVAS Intelligence is a chip design and solutions company that develops general AI computing technology. The company provides hardware acceleration through its EPOCH series of AI compute chips and PCIe compute cards, which are designed for large-model workloads and parallel computing.
The technology is aimed at enterprise companies and the automotive sector. It supports vehicle-based applications, including intelligent cockpit engines and central AI computing hubs that manage multi-domain data and decision-making.
Buyers should confirm whether the RISC-V and VISA architecture aligns with their existing software toolchains and if the hardware specifications meet the specific latency and power requirements of their environment.
Key Features
A fused architecture combining open RISC-V instruction sets with a proprietary virtual instruction set for AI workloads.
Processors featuring vector engines and tensor units designed for hardware-level acceleration of large AI models.
Hardware acceleration cards using PCIe 5.0 interfaces for integration into standard server infrastructure.
An AI solution for vehicles that supports image enhancement in real time and cabin sensing via DMS and OMS.
A vehicle AI box designed to replace multiple distributed ECUs with a centralized computing architecture.
Native support for FP8 and INT4 data types to help reduce inference latency and power consumption.
Use Cases
Using AI engines to support image quality enhancement in real time and driver/passenger monitoring.
Providing compute power for autonomous driving features and vehicle AI decision-making.
Deploying PCIe compute cards to accelerate the inference of large models in financial or telecommunications sectors.
Consolidating multiple vehicle control functions into a single AI-native central computing hub.
Best For
Integrations
Pricing
Pricing was not clearly available from the provided evidence. Buyers should confirm current pricing on the vendor website.
FAQ
EVAS uses a fused architecture combining the open RISC-V instruction set and its proprietary VISA virtual instruction set.
While it focuses heavily on the automotive industry for smart cockpits and driving, it also serves software and enterprise companies in finance and telecommunications.
Yes, the EPOCH PCIe compute cards are compatible with frameworks such as PyTorch and TensorFlow.
Source category: Software Development
Source subcategory: AI Infrastructure
