Thunder Compute

What it does

Thunder Compute provides cheap cloud GPUs for AI/ML, offering on-demand A100 & H100 GPU instances for up to 80% less. It simplifies AI/ML prototyping and development with tools like a simple CLI, instance templates, and VS Code/Cursor extension for real-time file syncing. Users can scale on demand and pay only for what they use.

How to use: Create an instance, search/install a model, and run it using templates like Ollama. Develop in VS Code with the provided extension for real-time file syncing.

Core features

• ✦On-demand GPU instances (A100, H100, Tesla T4)
• ✦Simple CLI for cloud development
• ✦Instance templates for common AI/ML tools
• ✦VS Code/Cursor extension for real-time file syncing
• ✦Pay-as-you-go pricing

Use cases

• →AI/ML prototyping and development
• →Training, fine-tuning, and deploying AI models
• →Self-hosting AI/ML applications
• →Transcribing video