Vulcan

Thank you I'll check and update here.

On a AMD EPYC 64 core 240 threads cloud instance it is stuck like this with 240 threads. I noticed that above a certain number of threads its slow, or...

So I have tried with the above mentioned cloud provider various number of threads. I found that anything above 64 threads gets slower and usable upto 120 threads. Anything above...

Env: Restricted Cloud / Throttled Maybe CPU: AMD EPYC 7742 64-Core Processor OS: ``` Distributor ID: Ubuntu Description: Ubuntu 20.04.3 LTS Release: 20.04 Codename: focal Linux XXXX 5.4.0-131-generic #147-Ubuntu SMP...

Okay, 8 threads max, so for a large file, is there a possibility of splitting the file to chunks with silences as terminators and dividing the conversion to ((total threads/cores)/8)...

> You can generate a table with performance results by simply running the [extra/bench_all.sh](https://github.com/ggerganov/whisper.cpp/blob/master/extra/bench-all.sh) script. Hey Sorry. That didn't pan out well, I did the benchmark thrice, my account got...

Here is my recommendation, for x86 windows and multiple targets at once in a single binary format with zero to minimal changes: https://github.com/jart/cosmopolitan For a demo of what is possible...

PS. If you go the cosmopolitan way, you could also think of using redbean to encapsulate it and provide a web ui - would be cool.

It is theoretically possible, but almost all uC boards lack the flash to store the models and have insufficient ram. You'd need to design a uC board with sufficient RAM/PSRAM...

@tkchia It would be great if you could test your patch by building this: https://github.com/trholding/llama2.c The corresponding issue is this: https://github.com/jart/cosmopolitan/issues/866 Read this to build: https://github.com/trholding/llama2.c#binary-portability-even-more-magic Model can be downloaded...