Angehefteter Tweet

Happy Sunday fam!
Using the @OpenGradient SDK stack, I built a small demo showing how an AI API call can become verifiable.
Every LLM prompt runs inside a Trusted Execution Environment, and each inference produces a payment receipt through x402.
Instead of simply trusting the output, the execution can be traced and verified.
Demo: Opengradient-vaod.xyz
With this demo, you get:
• The AI response
• Proof that the computation actually ran
• Execution metadata
• A payment receipt via x402
You can ask any question (Opengradient related too).
Prefer testing it with Claude Opus 4.6 but you can try other models.
Disclaimer: This is not a paid tweet, just complying with X rules.

English



































