yes it works with my Engine and makes silent payment possible

it works on my engine
https://github.com/shrec/UltrafastSecp256k1

I guess if you want to grind txids?

https://github.com/craigraw/bench_bip352 this Benchmark writen not by me it works on real workflow not just benchmark. craigraw write it for BIP352 based on my engine and i optimized it for him on gpu alos in my fork if interested all are reproducible with benchmarks and audits.

https://github.com/shrec/UltrafastSecp256k1

UltrafastSecp256k1 v3.21 released.

Major improvements:
• constant‑time SafeGCD scalar inverse
• faster CT ECDSA signing
• stricter BIP‑340 parsing
• expanded audit infrastructure

The project now includes cross‑platform benchmarks and reproducible Docker CI.

GitHub:
https://github.com/shrec/UltrafastSecp256k1

deleted by author

Thank you that was exactly the goal.
Portable, minimal, and easy to integrate across platforms.

I’m continuing the audit and cross-platform benchmarks feedback is very welcome.

ARM64 on android Device and RISC-V on Milk-V Mars

at this moment i have few esp32 devices one RISC-V Milk-V Mars and my desktop Intel and RT5060 ti GPU and i'm benchmarking and testing evrything on this devices and one STM32 mcu i don't have AMD cpu and GPU to test OpenCl on amd but works on Intel GPU without problem. Also i Have one Android Arm64 Device and testing ARM 64 on it but dont have RASPI and other ARM or AMD devices to do real benchmarks. on Apple Mettal helped friend with Macbook pro M3 i have it's benchmarks also in repo.