HackerTrans
トップ新着トレンドコメント過去質問紹介求人

steeve

2,229 カルマ登録 14 年前
http://twitter.com/steeve http://github.com/steeve

[ my public key: https://keybase.io/steeve; my proof: https://keybase.io/steeve/sigs/Jh6dVy9uklGwmVZTz6QflyRhA9l7BRA6VoZeuBK5SLo ]

投稿

ZML/LLMD alpha – cross platform LLM server

zml.ai
16 ポイント·投稿者 steeve·3 日前·3 コメント

Zml-smi: universal monitoring tool for GPUs, TPUs and NPUs

zml.ai
77 ポイント·投稿者 steeve·3 か月前·11 コメント

ZML/v2: Frontier performance through composability

zml.ai
5 ポイント·投稿者 steeve·4 か月前·0 コメント

コメント

steeve
·先月·議論
sidenote: it is possible to use Vulkan to map GPU memory to CPU space and even map it back to CUDA: https://x.com/steeve/status/2055042304344231978?s=20
steeve
·3 か月前·議論
consider it done
steeve
·3 か月前·議論
Weird, because we tried it. It doesn’t show anything?

We use the amdsmi to get metrics. I’ll investigate.
steeve
·3 か月前·議論
sadly, sandboxing is something that can't be upstreamed. this way, sandboxing is kept in zml instead of patching mesa.

as for nvtop, great program, but we missed a few features (such as sandboxing)
steeve
·4 か月前·議論
hi, i'm the founder of https://github.com/zml/zml, very happy with Zig
steeve
·4 か月前·議論
Misleading title. It's about using their AI service, not the editor.
steeve
·5 か月前·議論
we (ZML) have been back to following Zig master since std.Io was introduced. It's not that bad tbh. Also most changes really feel like actual improvements to the language on a day to day basis.
steeve
·5 か月前·議論
ZML
steeve
·6 か月前·議論
tldr: they replaced using protobuf as the type system across language boundaries for FFI with true FFI
steeve
·6 か月前·議論
i can't help but think of the scammy communication around this on X. and the actual need to be deceitful about it.

oh how low the mighty have fallen
steeve
·6 か月前·議論
very very cool !
steeve
·6 か月前·議論
This looks incredible, congratulations!
steeve
·6 か月前·議論
the upside is absolutely worth it
steeve
·6 か月前·議論
because he was been sending me updates about it for the last two weeks

because we've been discussing architecture, optimisations and benchmarks

because it's not a 13h old project
steeve
·6 か月前·議論
he’s been working on it for 2 weeks, as he said somewhere else
steeve
·6 か月前·議論
sure, i hereby disclose that he left in September

i already said in another reply that i knew him personally
steeve
·6 か月前·議論
the way he came barging through the door doesn’t invite sympathy
steeve
·6 か月前·議論
well, bazel is by far the most reliable one so I'm not sure why you're complaining
steeve
·6 か月前·議論
and it would be terrible for hermeticity and reproducibility, nix tries very hard and gets mediocre results

perhaps, just perhaps, why people go through the trouble not because they are idiots but for actual engineering reasons
steeve
·6 か月前·議論
[flagged]