Microsoft is set to unveil its in-house AI chip, codenamed "Athena," challenging NVIDIA's GPUs.
Athena aims to lessen Microsoft's dependence on NVIDIA-designed GPUs, facing supply constraints.
The chip's debut is expected at the Ignite conference, starting on November 14th in Seattle.
Athena is designed to compete with NVIDIA's flagship H100 GPU for AI acceleration in data centers.
Athena has undergone secretive testing by small groups within Microsoft and partner OpenAI.
Athena's development began in 2019 as Microsoft sought to cut costs and reduce reliance on NVIDIA.
NVIDIA GPUs are preferred by AI developers due to the CUDA platform, making attracting users to Athena crucial for Microsoft.
Microsoft aims to decrease reliance on NVIDIA, especially in the face of tight GPU supplies.
AI chips are pivotal in data centers. Microsoft competes fiercely, providing choices to cloud customers and forging an independent path in AI infrastructure.