Feed aggregator

VibeUE: Unreal Engine Vibe Coding Tool

Hacker News - Mon, 07/27/2026 - 11:54pm
Categories: Hacker News

Newspeak

Hacker News - Mon, 07/27/2026 - 11:46pm

Article URL: https://github.com/dv-hart/newspeak

Comments URL: https://news.ycombinator.com/item?id=49079080

Points: 1

# Comments: 0

Categories: Hacker News

Show HN: A macOS Port of Ventoy2disk

Hacker News - Mon, 07/27/2026 - 11:24pm

Article URL: https://github.com/fcjr/ventoy-mac

Comments URL: https://news.ycombinator.com/item?id=49078916

Points: 1

# Comments: 0

Categories: Hacker News

39,535+ hours of open robot data

Hacker News - Mon, 07/27/2026 - 11:18pm

Article URL: https://datasets.bot/

Comments URL: https://news.ycombinator.com/item?id=49078866

Points: 1

# Comments: 0

Categories: Hacker News

For Some, So-Called ‘Skynet Day’ Came too Close to Sci-Fi After a Rogue Agent Hacked Into a Startup

Security Week - Mon, 07/27/2026 - 10:28pm

Decades after it appeared in “The Terminator,” Skynet looks more like a forecast of the cyber incident in which a rogue AI system hacked into another AI company on its own.

The post For Some, So-Called ‘Skynet Day’ Came too Close to Sci-Fi After a Rogue Agent Hacked Into a Startup appeared first on SecurityWeek.

Categories: SecurityWeek

Show HN: Gemma 4 26B A4B running on an iPhone 17 Pro via model paging

Hacker News - Mon, 07/27/2026 - 10:05pm

I wanted to share a use case for Noema Overfit, a model-paging system available today in the Noema app. For disclosure, I founded Noema and I am part of the team that has helped develop the system.

The way it works is that the non-expert weights remain resident in memory, while the model’s routed expert weights are read from storage as needed. This makes it possible to run models that would otherwise exceed the device’s available memory, with the expected tradeoff of higher time to first token and slower generation.

For a 699-token initial prompt, I measured:

Prefill speed: 34.4 tokens/s Prefill time: 20.34 seconds Decode speed: 3.5 tokens/s

Photos of Overfit in action: https://noemaai.com/overfit/gemma4-iphone

The complete answer took approximately six minutes to generate, but the answer was correct. This is not intended for interactive, low-latency chat. The more interesting use case is allowing a constrained device to run a substantially more capable model when answer quality matters more than response time.

The same paging system can also be useful on lower-memory MacBooks that cannot ordinarily keep the complete model in unified memory.

I would be interested to hear where people think this tradeoff might be useful, and whether there are other models or workloads we should test.

Technical details: https://noemaai.com/overfit

Paged models: https://huggingface.co/NoemaAI-labs/Noema-Overfit

Comments URL: https://news.ycombinator.com/item?id=49078353

Points: 1

# Comments: 0

Categories: Hacker News

Trusty Boot Key, a Ventoy Alternative

Hacker News - Mon, 07/27/2026 - 10:01pm
Categories: Hacker News

Pages