frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Nvidia's $750B in Deals Reignite Circular AI Fears

https://www.bloomberg.com/news/articles/2026-07-27/nvidia-s-750-billion-deals-revive-fear-of-ai-c...
1•petethomas•38s ago•0 comments

A work queue that is one database table (ljms)

https://github.com/Anode1/ljms
1•siberean•53s ago•0 comments

How HN: ReturnFast – a global Back and Forward history for macOS

https://www.returnfa.st
1•andbalashov•2m ago•0 comments

Etsy Is in Its Flop Era, and Sellers Are Fleeing

https://www.wired.com/story/etsy-is-in-its-flop-era-and-sellers-are-fleeing/
2•cdrnsf•2m ago•0 comments

In China, people are renting out their faces to AI

https://restofworld.org/2026/china-ai-microdramas-face-licensing/
1•cdrnsf•4m ago•0 comments

Europe wildfires rage on, forcing more than 250k to flee

https://www.dw.com/en/europe-wildfires-rage-on-forcing-more-than-250000-from-homes/a-78109406
2•speckx•4m ago•0 comments

Show HN: A curated index of open-source AI developer tools

https://github.com/Sami-Uysal/awesome-open-ai-developer-tools
1•narheur•5m ago•0 comments

Wi-Fi 7 Router Buying Guide: Specs That Mean for Gaming and Streaming

https://capytoolkit.com/blog/buying-guides/wifi-7-router-buying-guide-specs-gaming-streaming/
1•CapyToolkit•5m ago•0 comments

All major LLMs are lib-left. Even Grok, half the time

https://unslop.run/blog/political-compass-of-llms
1•dopamine_daddy•5m ago•1 comments

Private companies could ruin the night sky and international law can't stop them

https://www.space.com/space-exploration/satellites/private-companies-could-ruin-the-night-sky-for...
1•Gigamouse•6m ago•0 comments

AI Is Coming for Every Bluff We Live By

https://michaelnovati.substack.com/p/ai-is-coming-for-every-bluff-we-live
1•gpi•6m ago•0 comments

Time Doesn't Trickle Down Either

https://thenoosphere.substack.com/p/time-doesnt-trickle-down-either
1•hn_acker•7m ago•0 comments

Segmented YouTube video cc –> .txt and .pdf

1•cristyg0101•9m ago•1 comments

An undo layer for coding agents that ship too fast

https://safecommit.vercel.app/
1•Shmungus•9m ago•0 comments

Cursor's actual Tab model in Neovim at zero setup

https://github.com/teocns/neocursor.nvim
1•DemiGuru•9m ago•0 comments

Show HN: Let's Seal – Let's Encrypt for document signing, free and self-hosted

https://github.com/letsseal/letsseal
1•nsokin•10m ago•0 comments

I Spent $50 and Five Days Proving a 2.7-Second Oracle Lead Is Worth Nothing

https://i-a-s.me/blog/polymarket-oracle-lead/
1•I-A-S•10m ago•0 comments

Where should your company's AI brain live?

https://iamcharliegraham.substack.com/p/where-should-your-companys-ai-brain
2•tylerg•13m ago•0 comments

Octane, the successor to Inferno – React's programming model, compiled

https://twitter.com/trueadm/status/2081764408741540071
3•steadicat•14m ago•0 comments

Bad Apple but It's Traceroute

https://jssfr.de/2026-07-27-bad-apple-but-traceroute.html
1•jssfr•14m ago•0 comments

Singapore's Unique Monetary Policy

https://www.reuters.com/world/asia-pacific/how-singapores-unique-monetary-policy-works-2026-07-27/
1•kaycebasques•14m ago•0 comments

Replacing gas stoves with electric stoves greatly reduces asthma symptoms

https://theconversation.com/replacing-gas-stoves-with-electric-stoves-greatly-reduces-asthma-symp...
1•speckx•14m ago•0 comments

The Billionaire Blowing the Whistle on Trump's Crypto Shop

https://nymag.com/intelligencer/article/justin-sun-world-liberty-lawsuit-trump-crypto-herro.html
1•cpncrunch•15m ago•1 comments

The Problem with Capitalism – Smarter Every Day [video]

https://www.youtube.com/watch?v=vS6HEes8daw
2•num42•16m ago•0 comments

Why do we have traveling brain waves?

https://www.salk.edu/news-release/why-do-we-have-traveling-brain-waves/
1•bookofjoe•16m ago•1 comments

Expanding Our Partnership with Cognizant

https://www.anthropic.com/news/cognizant-anthropic
1•surprisetalk•16m ago•0 comments

California's economy isn't just bigger than TX and FL – it's growing faster, too

https://www.gov.ca.gov/2026/07/24/californias-economy-isnt-just-bigger-than-texas-and-florida-its...
3•littlexsparkee•17m ago•1 comments

A toy Swift for the browser, in Rust

https://tswift.pages.dev/
1•CharlesW•18m ago•0 comments

How I drive a cloud coding agent from my phone through the browser

https://ykdojo.github.io/antigravity-cloud-run/posts/phone-access-iap.html
1•ykev•18m ago•0 comments

Kimi K3 on vLLM: Up to 370 Tokens/sec

https://vllm.ai/blog/2026-07-27-k3
1•wskwon•18m ago•0 comments
Open in hackernews

Kimi-K3 Technical Report [pdf]

https://github.com/MoonshotAI/Kimi-K3/blob/main/k3_tech_report.pdf
111•vinhnx•38m ago

Comments

m00dy•27m ago
I would want to see three things before drawing strong conclusions:

End-to-end tokens/sec and cost on realistic coding agent trajectories, including tool outputs and retries, not isolated decode benchmarks.

Cache hit rates and prefill cost for branching, multi-turn sessions.

Router-load distributions after post-training, where expert collapse or specialization problems often show up.

storus•22m ago
What would be the current best method to fine-tune it for my own specific agentic tasks? LoRA + DPO? GRPO? Something else?
whimsicalism•21m ago
LoRA + SFT, but it'll be big - better to wait for a finetuning API from one of the providers, I wouldn't jump straight to RL or off-policy pseudo-RL like DPO.
m_ke•17m ago
Also open sourced a bunch of infra to go with it.

Anyone who claims open source and open weights models are "decel" needs to get their head checked

https://github.com/MoonshotAI/MoonEP

https://github.com/kvcache-ai/AgentEnv

https://github.com/MoonshotAI/FlashKDA

jvanderbot•15m ago
This comment would be much better without the second line
ike_a•9m ago
I'm not sure I understand the case for open-source models being decelerationist, is this it?

Decel:

- Potentially reduces investor appetite for funding big labs.

- More risk of powerful AI getting in bad hands -> more regulation.

Accel:

- More competition so big labs can't rest on laurels.

- More research in open, so all labs can accrete advancements faster.

I feel like open-source = acceleration has a much more clear argument. (and how bad would deceleration be in any case?)

Iolaum•6m ago
Open Source models decelerate growth of closed AI. For people who think (or want) AI = closed_AI then that argument has weight. Good luck getting them to update their priors.
StevenWaterman•6m ago
I think it's basically open weights => more inference competition => less profit from inference => less training competition
acedTrex•8m ago
Why?
fahrradflucht•13m ago
License: https://huggingface.co/moonshotai/Kimi-K3/blob/main/LICENSE

> If the Licensee or any of its affiliates operates a Model as a Service business, and the aggregate revenue of the Licensee and its affiliates exceeds 20 million US dollars (or the equivalent in other currencies) in total over any consecutive 12 months, the Licensee must enter into a separate agreement with Moonshot AI before using the Software or its derivative works for any commercial purpose.

+ the existing 100 million monthly active users, or more than 20 million US dollars for commercial products have to name Kimi clause

throwaway27448•3m ago
I wonder how they'll figure out who to target for litigation when this license is violated.
NitpickLawyer•2m ago
Interesting license. They basically added a "no composer / better models" if revenues (including downstream) above 20m clause, where you'd have to get a separate agreement with them.

Still has the attribution clause if serving this to your users.

Other than that, free to do whatever you want.

whimsicalism•5m ago
It's funny that we've finally returned to tanh activation functions, time is a circle.
a-dub•2m ago
knowledge graph guided task synthesis. very cool! i have long wondered about the "how do you get good coverage of all the tasks" problem.