Hacker News

Article URL: https://api-first-world.com/

Comments URL: https://news.ycombinator.com/item?id=49027278

Points: 1

# Comments: 1

Article URL: https://www.foreignaffairs.com/china/chinas-moment-weakness

Comments URL: https://news.ycombinator.com/item?id=49027269

Points: 1

# Comments: 0

I’ve been building Echo (https://echo.tracerml.ai/), an experiment in making one AI system out of a pool of open-weight models rather than choosing a single model and using it for every task.

It started with a simple experiment. I took a group of models, including GLM-5.2, Kimi K2.7 and others, and ran them on the same evaluations. Then I measured what would happen if, for each problem, you somehow knew in advance which models would be useful and how their outputs should be combined.

That hypothetical system performed substantially better than any individual model in the pool. Of course, it is not something you can actually deploy because it relies on knowing which decisions were good after seeing the result. Echo is my attempt to recover some of that advantage without having that information in advance.

For each request, Echo decides how much computation to allocate, which models should participate, and how their work should be combined. Some prompts may only need a relatively small amount of inference, while others benefit from multiple models working on different parts of the problem.

One thing that surprised me while building it was how complementary the models are. A model that is clearly weaker overall can still be extremely useful on particular problems or as part of a combination.

On my first evaluation mix, Echo consistently performed better than the best individual model in its pool. It also reached roughly the same aggregate result as Fable, which I used as one of the stronger comparison systems, at around one third of the inference cost.

There are still some cases where Echo makes the wrong allocation or combination decision. I’m currently spending a lot of time understanding those failures, as well as testing whether the same approach holds up on coding and agentic tasks where measuring the quality of each decision becomes much harder.

I built a chat interface (echo.tracerml.ai) and an OpenAI-compatible API (https://echo.tracerml.ai/docs/api) so the system can be tested outside the evaluation setup.

Here is a short/high level video on how it works: https://youtu.be/lJFJSvOdXhg

I wrote up the evaluation methodology, individual model results, costs and current limitations here: https://echo.tracerml.ai/eval

I would love for you to try it! Especially if you hit any weird failure cases or places where the allocation looks unintuitive.

Comments URL: https://news.ycombinator.com/item?id=49026810

Points: 1

# Comments: 0

Article URL: https://www.cnn.com/2026/07/23/politics/republicans-capitalism-socialism-poll

Comments URL: https://news.ycombinator.com/item?id=49026787

Points: 2

# Comments: 0

Hi guys,

Just needed a good way to quickly record a beautiful product demo on Windows with automatic zoom, cursor highlights, background etc.

Tried a couple of products but they pissed me off (heavy, slow, expensive, didn't let you use all the features).

So I built this in anger. It's simple, lightweight, free (and always will be), and creates beautiful videos.

Grateful for any issues or PRs.

Thanks, Chris

Comments URL: https://news.ycombinator.com/item?id=49026781

Points: 1

# Comments: 0

Article URL: https://www.msn.com/en-us/news/other/media-silence-as-kushner-s-2b-tie-to-saudis-slips-under-cover-of-trump-deal-analyst/ar-AA28xRF6

Comments URL: https://news.ycombinator.com/item?id=49026757

Points: 2

# Comments: 0

Article URL: https://www.nrl.navy.mil/Media/News/Article/4551871/robotic-servicing-of-geosynchronous-satellites-launches-into-orbit/

Comments URL: https://news.ycombinator.com/item?id=49026755

Points: 2

# Comments: 0

Article URL: https://www.washingtonpost.com/health/2026/07/23/fda-advisory-panel-vote-recommending-7-peptides-broader-access/

Comments URL: https://news.ycombinator.com/item?id=49026749

Points: 1

# Comments: 1

Article URL: https://www.independent.co.uk/news/science/bryan-johnson-clone-baby-biohacking-b3020197.html

Comments URL: https://news.ycombinator.com/item?id=49026734

Points: 3

# Comments: 0

Article URL: https://www.theguardian.com/environment/2026/jul/23/orcas-ramming-prey-hard-explodes-playing-game

Comments URL: https://news.ycombinator.com/item?id=49026693

Points: 3

# Comments: 0

Avery Software

Comments URL: https://news.ycombinator.com/item?id=49026686

Points: 1

# Comments: 0

Article URL: https://cupqualifier.app

Comments URL: https://news.ycombinator.com/item?id=49026681

Points: 1

# Comments: 0

Article URL: https://tris.sherliker.net/blog/dont-put-your-name-to-bot-written-content/

Comments URL: https://news.ycombinator.com/item?id=49026669

Points: 3

# Comments: 0

Article URL: https://gammavibe.com/updates/i-built-a-premium-course-and-sold-zero-copies/

Comments URL: https://news.ycombinator.com/item?id=49026666

Points: 2

# Comments: 0

Article URL: https://economist.com/finance-and-economics/2026/07/23/the-rise-of-the-deserving-rich

Comments URL: https://news.ycombinator.com/item?id=49026639

Points: 3

# Comments: 0

Article URL: https://www.playmagicreef.com

Comments URL: https://news.ycombinator.com/item?id=49026606

Points: 2

# Comments: 1

Article URL: https://pwno.io/diff

Comments URL: https://news.ycombinator.com/item?id=49026605

Points: 3

# Comments: 0

Article URL: https://arstechnica.com/space/2026/07/the-space-force-is-now-seeking-to-buy-up-to-30-billion-in-rocket-launches/

Comments URL: https://news.ycombinator.com/item?id=49026602

Points: 3

# Comments: 0

Article URL: https://github.com/arizqi/cpubrrr

Comments URL: https://news.ycombinator.com/item?id=49026595

Points: 1

# Comments: 0

Article URL: https://leetcopilot.dev/blog/how-ai-chat-changed-my-leetcode-practice

Comments URL: https://news.ycombinator.com/item?id=49026567

Points: 2

# Comments: 1