World Train Map – 1247 train routes around the world
Article URL: https://worldtrainmap.com/ Comments URL: https://news.ycombinator.com/item?id=49250894 Points: 6 # Comments: 0
Latest
b10354
ggml-cpu : fix CPU affinity mask being ignored on Android (#26838) Website: https://llama.app macOS/iOS: macOS Apple Silicon (arm64) macOS Apple Silicon (arm64, KleidiAI enabled) DISABLED macOS…
RT Francesco: Re You can now try offline computer use with cua-driver + Muse Glimmer, through our friends at @ollama 🦙 How-to: https://cua.ai/docs/…
RT FrancescoRe You can now try offline computer use with cua-driver + Muse Glimmer, through our friends at @ollama 🦙How-to:https://cua.ai/docs/how-to-guides/driver/run-with-local-model
Google Search Is Dying. What Comes Next Is Worse
Article URL: https://thewalrus.ca/google-search-is-dying/ Comments URL: https://news.ycombinator.com/item?id=49250836 Points: 10 # Comments: 2
RT Thariq: Re @simonw we're working on removing Haiku from WebFetch now that automode is default
RT ThariqRe @simonw we're working on removing Haiku from WebFetch now that automode is default
Mark Zuckerberg's Yacht Refused to Help Ship in Distress
Article URL: https://nymag.com/intelligencer/article/zuckerbergs-yacht-booed-over-failing-to-help-stranded-boat.html Comments URL: https://news.ycombinator.com/item?id=49250731 Points: 14 # Comments: 1
Muse Glimmer on 1/2 AMD v620
Hey. Just tried it on my old ass gpus 😄 Surprisingly Tensor Split is working on 2 gpus almost doubling PP (wonder how it will work…
RT Arthur MacWaters: SF spends more than $1,000,000,000/yr now on homelessness (5x increase in 5 years) A wild number. And yet the homeless population…
RT Arthur MacWatersSF spends more than $1,000,000,000/yr now on homelessness (5x increase in 5 years)A wild number. And yet the homeless population has only gotten largerBecause…
RT Lulu Cheng Meservey: This Meta campaign is a case study in strategic reframing, with 4 major examples: 1) REFRAMES THE AI RACE FROM “who builds it…
RT Lulu Cheng MeserveyThis Meta campaign is a case study in strategic reframing, with 4 major examples:1) REFRAMES THE AI RACE FROM “who builds it the…
Ling 3.0 Flash on Strix Halo
vLLM ROCm/HiP, 4 bit compressed-tensors (int4) Not a fair comparison, but Qwen-122b on the most optimized format possible I have run (rocmFP4) does not touch Ling…
With new open models, Meta pitches another reboot of its struggling AI strategy
Meta has been trailing competitors. Zuckerberg thinks he's found a way forward.
Show HN: AI Pulse a fake LED strip beside the macOS Dock that shows agent status
Hi HN!I run a few Claude Code sessions in parallel and kept cmd-tabbing around just to find out one of them had been sitting on a…
Q: Is dual-use alignment-complete problem?
Personally, I believe it would be helpful for the alignment community to somehow quantify how much of a given piece of research goes directly into alignment…
CLIP-CC-Bench: Evaluating Paragraph-Level Video Descriptions in Video-Language Models
Benchmarking video-language models has largely focused on short clips and single-sentence metrics, leaving open whether current systems can generate accurate long-form, paragraph-level descriptions. We introduce CLIP-CC-Bench,…
b10353
ggml : require contiguous src for ROLL on CUDA and Metal (#25928) ggml_roll only asserts nb[0] == ggml_type_size, so a permuted src is a valid input,…
DuplexGen: Adaptive Synthesis of Human-AI Turn-Taking Dialogues
Turn-taking is a central component of full-duplex interaction. Which turn-taking behaviors are appropriate varies with the scenario, yet current models apply a single norm regardless of…
Mark Zuckerberg doesn’t understand how to live
Recently, a man I was rock climbing with told me about how he'd used AI to make a motivational poster for himself, which he'd hung on…
Forking-Sequences — Part II: Multi-Horizon Forecast Ensembling with Reduced Volatility
Based on: Potosnak, W., Wolff, M., Cao, M., Ma, R., Konstantinova, T., Efimov, D., Mahoney, M.W., Oreshkin, B., & Olivares, K.G. "Forking-Sequences: Statistically and Computationally Efficient…
The Pacing of the Frontier
In the wake of the letter calling on us to prepare to potentially Pace the Frontier, there has been much discussion of when pacing the frontier…
Claude Haiku is my current least favorite model – it hallucinates wildly, and is out-performed now by other similarly priced models like GPT-5.6-Luna …
Claude Haiku is my current least favorite model - it hallucinates wildly, and is out-performed now by other similarly priced models like GPT-5.6-LunaEven worse: it seems…
I trained a 1B-parameter LLM from scratch on 20B tokens for about $200
A few months ago, I had the idea of making a LLM from scratch as a personal project (for learning and partly for improving my resume).…
RT Paolo Rosson: Got Meta's new Muse Glimmer 30B running on my MacBook (M3 Max, 96GG) and tested the serving options available so far. Fastest right n…
RT Paolo RossonGot Meta's new Muse Glimmer 30B running on my MacBook (M3 Max, 96GG) and tested the serving options available so far.Fastest right now: Ollama's…
How Claude marks AI-generated content
Article URL: https://support.claude.com/en/articles/16266773-how-claude-marks-ai-generated-content Comments URL: https://news.ycombinator.com/item?id=49250109 Points: 14 # Comments: 5
We cut a 40-day financial integration down to 5 days using Google Anti
Article URL: https://discuss.google.dev/t/trusted-automation-with-google-antigravity-scaling-secure-finance-integrations-from-40-days-to-5/383313 Comments URL: https://news.ycombinator.com/item?id=49249986 Points: 3 # Comments: 0