News for hackers. Stuff that matters.

bbstats

Re:Artificial Analysis v5 and the benchmarker's dilemma on Friday October 30, 2026 @12:25PM (#50306219)
they do
this is built into teams
Re:You Are No Longer Invited to Dinner on Tuesday September 29, 2026 @09:35AM (#49893015)
Don't take a percent change of a percent change.
Re:AI-generated posters don’t have to be horrible on Saturday September 19, 2026 @09:07AM (#49766287)
the biggest issue is NO WHITE SPACE.
Re:The Two MMLU Scores: What a Benchmark Name Does Not Fix on Tuesday September 15, 2026 @09:18PM (#49721038)
"What a ____ does not ____" is painfully Claude-Coded (no pun intended)
this is great for the 'make a webapp for me and my family' thing
Re:Claude, change the "Add to Cart" button to blue on Wednesday September 09, 2026 @10:46AM (#49627501)
Mine immediately did it correctly?
Re:Gemini 3.8 Flash on Thursday September 03, 2026 @11:47AM (#49551969)
playing around with 3.8 - it generates non-working code. basically unusable.
Re:Pre-Release of Polars 2.0 on Thursday September 03, 2026 @11:21AM (#49551501)
Bothering me like crazy that "Use instead: .cat.to(dtype) for int → categorical, .cat.physical() for categorical → int." doesn't give the requisite code example!
Re:Karpathy's Pelican on Sunday August 02, 2026 @01:51PM (#49146651)
This is awful
Re:A Silly Adaptation of "The Odyssey" on Friday July 17, 2026 @08:16AM (#48946444)
Love to watch movies like this sad reviewer
Re:Inkling: Our Open-Weights Model on Wednesday July 15, 2026 @03:05PM (#48925606)
too bad we'll never know how good it is, since they used a radar plot to show its benchmark scores!
Re:MAI-Code-1-Flash on Tuesday June 02, 2026 @07:38PM (#48377723)
What application/UI are you using deep seek flash high on? Still copilot or something else
Re: on Thursday May 14, 2026 @07:00AM (#48133698)
The logic for which models stay active when you click on a group of them is extremely not working
First thing I searched was Pokemon cards and found items with bids at 50% higher than market value...either shill bidding or folks who are bidding blindly.
Re:Zed, A sans for the needs of 21st century on Monday April 13, 2026 @08:12AM (#47750882)
I would pay $8 for this
the subheadings are extremely AI IMHO
I mean...this is a huge company. For them to do something even mildly consumer-focused I'll take as a win.
their green screen is good. that has nothing to do with this video.
... did you watch the video?
inference is not usually slow for classical ML (save lightgbm)
Re:I baked a pie every day for a year and it changed my life on Thursday February 26, 2026 @04:17PM (#47172113)
Eating enough pie could help with that
Re:Half million 'Words with Spaces' missing from dictionaries on Wednesday February 25, 2026 @10:47AM (#47153124)
these are called phrases
all this because of a single week?
would love a multi-image version of this.
Re:Transparent Leadership Beats Servant Leadership on Thursday December 04, 2025 @10:01AM (#46148401)
this is so silly. porque no los dos?
Re:YesNotice on Tuesday December 02, 2025 @10:59AM (#46122550)
hug of death?
did you read the article?
"previously people could cheat, now they can't" ?? how is that bad.

Older »