NEAR AI Cloud enables private inference for GitHub Copilot

NEAR AI Cloud has introduced a method to connect its models to GitHub Copilot Chat via the ‘More Providers’ VS Code extension. This integration leverages NEAR AI Cloud’s OpenAI-compatible API, allowing developers to utilize external models within their existing workflows without significant reconfiguration.

The service emphasizes privacy through the use of Trusted Execution Environments (TEEs) powered by Intel TDX and NVIDIA confidential computing technologies. This setup is designed to isolate prompts, model weights, and outputs from infrastructure providers and the model owners themselves. Each inference request includes a hardware-signed cryptographic attestation to verify that the computation occurred within a secure environment.

Users can access various open-weight models, such as GLM 5.3 Flash, with a pricing structure based on token usage. Notably, the service offers a crypto-native payment option where users can stake NEAR tokens to generate inference credits, allowing them to maintain their original stake while utilizing the service.