fp.
newest
Open in hackernews
Mistral releases updates to Magistral family, adds vision
https://twitter.com/MistralAI/status/1968670593412190381
1
•
pember
•
1h ago
Comments
incomingpain
•
1h ago
I tested this yesterday for coding. VERY good model for coding. Crazy that small 1.2 is better than medium 1.1.
Certainly on par with qwen3 30b thinking in how well it did.
Nepal Gen-Z Revolution
https://www.theguardian.com/world/2025/sep/10/nepal-gen-z-protests-corruption
2
•
shrubby
•
1m ago
•
0 comments
iOS 26's Liquid Glass Design Draws Criticism from Users
https://www.macrumors.com/2025/09/17/ios-26-liquid-glass-critiques/
1
•
wslh
•
2m ago
•
1 comments
When Justice Holds a Knife: reflections on the death penalty's irreversible edge
https://www.clickworlddaily.com/2025/09/when-justice-holds-knife-reflections-on.html
1
•
jonalgarve
•
3m ago
•
1 comments
Cladh Hallan
https://en.wikipedia.org/wiki/Cladh_Hallan
1
•
bediger4000
•
3m ago
•
0 comments
First large health language model predicts >1000 diseases for a person and when
https://erictopol.substack.com/p/dawn-of-a-new-era-of-primary-prevention
1
•
ck2
•
3m ago
•
1 comments
Secrets of DeepSeek AI model revealed in landmark paper
https://www.nature.com/articles/d41586-025-03015-6
2
•
bookofjoe
•
4m ago
•
1 comments
Why 'false authenticity' is so unsettling
https://psyche.co/notes-to-self/why-walking-through-historic-wurzburg-is-so-unsettling
1
•
herbertl
•
4m ago
•
0 comments
Why Your AI Code Bias Is Making You a Worse Reviewer
https://revelry.co/insights/artificial-intelligence/your-ai-code-bias-makes-you-a-worse-reviewer/
2
•
sighren
•
4m ago
•
0 comments
Discovering new solutions to century-old problems in fluid dynamics
https://deepmind.google/discover/blog/discovering-new-solutions-to-century-old-problems-in-fluid-...
1
•
roboboffin
•
5m ago
•
0 comments
Unlock one-of-a-kind shows in unique spaces
https://www.sofarsounds.com/
1
•
fcpguru
•
5m ago
•
0 comments
Feature Flags in Depth
https://rtpg.co/2023/11/15/feature-flags-in-depth/
1
•
rbanffy
•
6m ago
•
0 comments
Show HN: I wrote an OS in 1000 lines of Zig
https://github.com/botirk38/OS-1000-lines-zig
1
•
botirk
•
10m ago
•
1 comments
The Worldcoin Scam Just Got Worse [video]
https://www.youtube.com/watch?v=lAaIRWjKAmw
2
•
askl
•
10m ago
•
0 comments
Apple's Audio Labs
https://www.techradar.com/audio/earbuds-airpods/apple-took-me-into-one-of-the-quietest-rooms-anyw...
2
•
coloneltcb
•
10m ago
•
0 comments
Ask HN: Parental control for a 13yo, on Linux (remote, web-based, extensible)
1
•
markun
•
11m ago
•
0 comments
KDE offers a cool Mobile UI, too
https://plasma-mobile.org
1
•
eloeffler
•
13m ago
•
0 comments
Elasticsearch Was Never a Database
https://www.paradedb.com/blog/elasticsearch-was-never-a-database
4
•
todsacerdoti
•
14m ago
•
1 comments
Reactive carbon runoff from logged forests undermines climate mitigation
https://phys.org/news/2025-09-soil-runoff-forests-reactive-carbon.html
2
•
PaulHoule
•
14m ago
•
0 comments
US-UK pact will create jobs
https://www.gov.uk/government/news/us-uk-pact-will-boost-advances-in-drug-discovery-create-tens-o...
2
•
monkeydust
•
15m ago
•
0 comments
Japan to investigate undersea-cable supply chains for China exposure Asia
https://asia.nikkei.com/business/telecommunication/japan-to-investigate-undersea-cable-supply-cha...
2
•
rbanffy
•
16m ago
•
0 comments
Using GPT-5 to prove new theorems on matrix multiplication
https://rybindmitry.github.io/blogs/gpt5-matrix-multiplication.html
1
•
frozenseven
•
17m ago
•
0 comments
Erlang-Red – Interview with Gerrit Riessen at TADSummit
https://blog.tadsummit.com/2025/09/17/erlang-red/
1
•
vkatsuba
•
17m ago
•
0 comments
Google kept its lobbying on this privacy bill quiet
https://themarkup.org/privacy/2025/09/12/google-wasnt-against-this-privacy-bill-officially-behind...
3
•
eustoria
•
20m ago
•
0 comments
Humans still better than AI for hotdog or not
https://www.humanprotocol.org/blog/evaluating-google-cloud-vision-for-image-moderation-how-reliab...
4
•
alexnewman
•
20m ago
•
0 comments
Signals in the Fire: The Impact of Banning Global Tech in Nepal
https://schmud.de/posts/2025-09-18-signs-in-the-fire-kathmandu.html
1
•
schmudde
•
22m ago
•
1 comments
China-backed cyber crew spoof Congressman to dig for dirt on US trade talks
https://www.theregister.com/2025/09/18/pandamonium_chinabacked_attackers_spoof_congressman/
2
•
rntn
•
22m ago
•
0 comments
ErnieApp – Privacy Knowledge Manager
https://ernieapp.com/
1
•
eustoria
•
23m ago
•
0 comments
A blog about the Cedar Policy Language
https://cedarland.blog/
1
•
mooreds
•
27m ago
•
2 comments
What did your pre-launch site look like?
1
•
not_your_mentat
•
27m ago
•
0 comments
The Larger a Company Gets, the Harder It Is for Anyone to Know What's Happening
https://twitter.com/hnshah/status/1968393736196624521
1
•
westi
•
28m ago
•
0 comments
Load More
incomingpain•1h ago
Certainly on par with qwen3 30b thinking in how well it did.