Dominic Pasceri

[Anthropic @AnthropicAI]

Anthropic has confidentially submitted a draft S-1 registration statement to the Securities and Exchange Commission.

Pending completion of SEC review, this gives us the option to pursue an initial public offering.

Read more: https://t.co/onGZAhRLvD

[vLLM @vllm_project]

Thanks to the community report, we recently identified a PR https://t.co/QWboSmskkF that attempted to solve a non-existent issue and was submitted as part of a “PR training” workflow for resume building.

The contributor involved has been banned from the vLLM community.

This kind of low-signal contribution increases maintainer review overhead and creates unnecessary operational costs for open-source projects.

As AI coding agents make generating large volumes of small PRs increasingly cheap, open-source communities will need to explore new ways to preserve contribution quality and reviewer trust.

While we are investigating how to deal with AI slop, we continue to highly value contributions from real users solving real production problems.

If you have an important contribution that has not yet received maintainer attention, please email us at:

[email protected]

Using a verifiable company or university email, include:
- your production or research use case
- the problem you encountered
- how your contribution addresses it

This helps us better prioritize impactful contributions while keeping the vLLM community open and collaborative.

As AI makes virtual contributors look increasingly real, authentic human collaboration matters more than ever.

vLLM’s mission remains unchanged: to make LLM inference easy, fast, and cheap for everyone — and we will continue working toward that goal.

[Dominic Pasceri @DP_SMH]

MLPerf training v6.0 key dates

submission UI opens: may 4
submission deadline: may 15
review period: may 18 to june 4
embargo lift: june 16

[Dominic Pasceri @DP_SMH]

@AnushElangovan you’re responding to a bot…

[Dominic Pasceri @DP_SMH]

@jukan05 around 8:15 onwards

https://t.co/3UBWrS3OV9

[Dominic Pasceri @DP_SMH]

gonna go out on a limb here and say that 60B option will not be picked up

[SpaceX @SpaceX]

SpaceXAI and @cursor_ai are now working closely together to create the world’s best coding and knowledge work AI.

The combination of Cursor’s leading product and distribution to expert software engineers with SpaceX’s million H100 equivalent Colossus training supercomputer will allow us to build the world’s most useful models.

Cursor has also given SpaceX the right to acquire Cursor later this year for $60 billion or pay $10 billion for our work together.

[Dominic Pasceri @DP_SMH]

@jukan05 you think he does it out of the kindness of his heart? the newsletter arguably the most valuable asset they have and it has nothing to do with revenue, it’s control of the narrative

[Hot Aisle @HotAisle]

ROCm 7.2.2

https://t.co/BNIypL5hku

[Dominic Pasceri @DP_SMH]

https://t.co/50ntS2xMXk

[Dominic Pasceri @DP_SMH]

ROCm skipped tests on pytorch went from around 60 on friday to 170 today

[Dominic Pasceri @DP_SMH]

@__tinygrad__ @Stefano02545325 @AMD @AnushElangovan @LisaSu yeah, i think a lot of people are gonna be upset when they spin it up and realize what they actually got for $1000.

[Dominic Pasceri @DP_SMH]

makes sense, the only reason to buy a B70 or any other “pro” card is if $INTC stock paid for it

[Dominic Pasceri @DP_SMH]

i remember getting so excited for the B60 hoping they’d price it under $500. silly me

[Dominic Pasceri @DP_SMH]

makes sense, the only reason to buy a B70 or any other “pro” card is if $INTC stock paid for it

[DVB @DeepValueBagger]

I'm celebrating @intel Terafab news by getting the Intel ARC Pro B70. The DDR6 has twice the memory bandwidth of a DGX Spark, and more than half price the price of the next Nvidia card.

Intel Arc Pro B70
32GB GDDR6 · 608 GB/s · 367 INT8 TOPS · 230W · oneAPI/SYCL · Soft ECC
$949

NVIDIA RTX 4090
24GB GDDR6X · 1,008 GB/s · 82.6 TFLOPS FP32 · 450W · CUDA · No ECC
~$2,755 new

[Dominic Pasceri @DP_SMH]

@dhh #ad

[Dominic Pasceri @DP_SMH]

@apaz_cli hate to see it 😔

[Dominic Pasceri @DP_SMH]

@TheWestbrookEra @steppingbacks @TrillzTerrell russ checked back in with 5 mins left in the 4th and went 0-5. every field goal they had after russ subbed in harden scored himself or assisted. he also had the game winning block on dort. what game did you watch?

[Dominic Pasceri @DP_SMH]

@jukan05 i think i still have some b die kits laying around

[Dominic Pasceri @DP_SMH]

7.12 dropped the day after i tweeted this.
technically one day late, but more significantly it marks the first parallel release since the inception of the preview stream

[Dominic Pasceri @DP_SMH]

if $AMD followed their own docs, ROCm 7.12 preview should’ve been releasing today (exactly 6 weeks after 7.11). instead they shipped 7.2.1 production. these streams were supposed to run in parallel. this is alternating.

[Anush Elangovan @AnushElangovan]

@DP_SMH @HotAisle @Stefano02545325 @AIatAMD The 7.9+ releases have been tagged preview but is on a regular cadence: https://t.co/ryGtH5pkPP

Since hipthreads is not part of regular ROCm release BOM I need to check if it has landed the fixes for the next release.

[Dominic Pasceri @DP_SMH]

INCOMING!