HackerTrans
TopNewTrendsCommentsPastAskShowJobs

darkolorin

no profile record

Submissions

Show HN: We made LM studio alternative based on own engine

trymirai.com
3 points·by darkolorin·il y a 9 mois·0 comments

Show HN: We made our own inference engine for Apple Silicon

github.com
186 points·by darkolorin·il y a 12 mois·46 comments

MBA Harvard Feels Undervalued

old.reddit.com
8 points·by darkolorin·l’année dernière·1 comments

Faster Inference Than MLX

trymirai.com
1 points·by darkolorin·l’année dernière·0 comments

90T/s on my iPhone llama3.2-1B-fp16

reddit.com
5 points·by darkolorin·l’année dernière·1 comments

comments

darkolorin
·il y a 12 mois·discuss
Basically “faster” means better performance e.g. tokens/s without loosing quality (benchmarks scores for models). So when we say faster we provide more tokens per second than llama cpp. That means we effectively utilize hardware API available (for example we wrote our own kernels) to perform better.
darkolorin
·l’année dernière·discuss
I made it! 90 t/s on my iPhone with llama1b fp16

We completely rewrite the inference engine and did some tricks. This is a summarization with llama 3.2 1b float16. So most of the times we do much faster than MLX. lmk in comments if you wanna test the inference and I’ll post a link.