Hacker News Reader

sourcecodeplzon When did Google get so weird?

i wrote about something similar here, the AI overview are doing REAL DAMAGE.

https://news.ycombinator.com/item?id=48717522

sourcecodeplzon Meta's Muse appears to use an OpenAI model labeled muse-special

zuck wouldn't do this, he is too proud imho

sourcecodeplzon Google’s Project Suncatcher to put ML infrastructure in space

oh forgot about ars. a very cynical crowd

sourcecodeplzon Google’s Project Suncatcher to put ML infrastructure in space

what is it exactly? jealousy i think. plus then add Elon hate. but now Google wants to do it too.

i mean why wouldn;t you want sci-fi things taking shape?

sourcecodeplzon Google’s Project Suncatcher to put ML infrastructure in space

no you didnt, i read most of your comments in this thread. very valuable even if little optimistic.

>> The current plan is for 250 kW peak with 160 m^2 radiators. It will weigh about 4 tons so you can pack 25 on one flight of Starship.

>> Deploy 4,000 and you're at 1 GW. That's 160 launches.

sourcecodeplzon Google’s Project Suncatcher to put ML infrastructure in space

> if you believe Elon.

Why wouldn;t you? he delivered on almost everything he promised, just the timeline was a lot later.

we are talking sci-fi things here, these will take time but can be done.

would you rather we bury our heads in the sand and never innovated beyond basics?

sourcecodeplzon Google’s Project Suncatcher to put ML infrastructure in space

just to add a few more reasons not mentioned on X why they chose this place:

Cooling is a major factor. Ulanqab's average temperature is about 4.3°C, enabling free air cooling nearly 10 months a year and very low PUE around 1.1.

Even more decisive is cheap, abundant power—largely wind and solar plus coal—at roughly half the price of eastern China. Add vast open land, dedicated low-latency fiber to Beijing (~2-5 ms),

sourcecodeplzon Grok 4.7

muse spark 1.3 contribs cache read is $0.002 btw (~220x diff).

sourcecodeplzon Grok 4.7

Output tokens from Intelligence Index:

- grok 4.6 (xhigh): 97M (for 44 score)

- grok 4.7 (xhigh): 240M (for 46 score)

sourcecodeplzon Grok 4.7

looks like token efficient/verbosity took a big hit.

Output tokens from Intelligence Index:

- grok 4.6 (xhigh): 97M (for 44 score)

- grok 4.7 (xhigh): 240M (for 46 score)

sourcecodeplzon Show HN: Local-coder – Build a team of coding agents with local models

nice project.

some benchmarks would be nice.

also consider adding some even smaller models too: minicpm, lfm

sourcecodeplzon Minimal Phone 2

minorities are the loudest.

truth is its quite difficult to function today without a smartphone.

banks, uber, food etc etc

edit: remember iphone mini? so "many" asked for it and apple finally made it. it was a flop.

sourcecodeplzon Show HN: Cactus Needle 3: 8-29MB automation models can match DeepSeek V4 Flash

from my limited tests, it can work with up to 10 tools/definitions.

over that and it gets confused

sourcecodeplzon Introducing System One Models and Jev

this reminds me of laravel boost, which does something similar.

it can generate classes/models/routers etc via tool calls, doesnt write the actual code.

sourcecodeplzon Mistral X Mozilla: Private, Multilingual AI Browsing

Seems like what Chrome has with the default built-in Gemini Nano model.

Hey at least "some" news/things from Mistral. Seems like ages ago when they launched vibe-code.

> Powers context-aware search, page summaries, and memory retrieval across browser tabs.

> Live in France and North America, with launches in the UK and Germany planned for later this year.

> Built on a zero data retention policy; conversations are not saved on Mozilla’s servers by default.

> Models are fine-tuned on regional languages, dialects, and local nuances to provide native-feeling responses.

> Extends Mistral’s open-weight AI models directly to consumer web browsing alongside its enterprise offerings.

sourcecodeplzon XCancel service is suspended until further notice

wtf the guy payed what 40billion for this website?

yes, he can enforce whatever ToS he wants .

sourcecodeplzon google.com/goto: Google's anti-scraping update

i tried visiting kagi now, and there is no way to search if you dont signup first?

also this page doesnt have a submit button and pressing enter doesnt submit the form either:

https://kagi.com/html

sourcecodeplzon Ask HN: What default model do you use and why?

i can not believe you really use Haiku?

sourcecodeplzon Ask HN: What default model do you use and why?

i like Luna but only until 130k context, over that it becomes dumb.

sourcecodeplzon Ask HN: What default model do you use and why?

Muse Spark 1.3 Contribs

unbeatable price/intel ratio per M tokens:

$0.10 (input)

$0.20 (output)

$0.002 (cached-input)

sourcecodeplzon Muse – Meta’s personal AI agent

> Sure, it can't book a flight or organize the family calendar for the next month, but I don't have any need for those.

why can't it do that? just run a VPS inside WSL or another container inside your PC. give the agent that sandbox.

sourcecodeplzon Nitter and XCancel resume service after legal advice

different things: this is an online service, not downloadable software.

sourcecodeplzon Show HN: Prices.eu.org

added some charts.

sourcecodeplzon Show HN: Prices.eu.org

Got it. Taxes % will be added next. Good idea thanks!

Also, even though the food prices aren't exact they are close. I think I should keep them.

sourcecodeplzon Show HN: Prices.eu.org

for sure more mining can be done for more sources. this is a first version and i plan on continuing to work on this site.

thanks for the feedback

sourcecodeplzon Show HN: Prices.eu.org

the pages are static html generated every day.

also there is no google analytics, fb pixel, cookies etc.

just a clean, fast site that updates its pages every day from official sources.

sourcecodeplzon Muse Spark 1.3

i've been using muse spark 1.2 contribs since launch exclusively. no other models.

i like it very much. it is different than all other chinese models distilled from claude.

just ask it to do some front-end work and you will see its not the same UI as all other claude/distils.

also the price is unbeatable, $0.002 input caching. its the same as old dsv4-flash prices.

sourcecodeplzon Muse Spark 1.3

lol of all people you think the Chinese will not log your queries?

sourcecodeplzon Muse Spark 1.3

they key pricing is cache reads at $0.002 per M (same as old deepseeek v4 flash prices)

sourcecodeplzon Muse Spark 1.3

price is same, also for contrib version