#dstack private-ai-gateway can now route to custom @chutes_ai deployments. The interesting part isn’t the routing. Before any prompt is forwarded, the gateway verifies the target GPU TEE against the expected measurements. And then it sends all the prompts via an e2e encryption.

View the Full episode here:

"Randomly export controlled on a Friday night." That is how fast access to a frontier model can vanish. No vote, no warning. One policy decision and the tool you built on is gone. @jon_durbin on the fix: sovereign models owned by the network. No export controls, no regional…

chutes.ai?utm_campaign=tk_bc… Link Chutes | Serverless AI Compute Deploy, run and scale any AI model in seconds. Try directly through our platform, or use our easy-to-use API in seconds. chutes.ai

AI servers will burn 175 TWh of electricity in 2026, up from 95 TWh last year. That is Gartner's forecast, and it is nearly double in a single year. The industry's answer is more gigawatts. Ours is fewer. Parallax trained a 20B model on cheap, mismatched GPUs to within about…