News for hackers. Stuff that matters.
Re:Show HN: Echo – Fable-level results at 1/3 the cost using... by dluan on Thursday July 23, 2026 @04:56PM (#49027950)
So this is the dogpile.com of the askjeeves, alta vista, and lycos approach? Time is a flat circle?
Re:Show HN: Echo – Fable-level results at 1/3 the cost using... by glasss on Thursday July 23, 2026 @05:26PM (#49028275)
Good ideas are usually still good across time and tools
Re:Show HN: Echo – Fable-level results at 1/3 the cost using... by onlyrealcuzzo on Thursday July 23, 2026 @06:44PM (#49029072)
Dogpile was only a good idea while Search Engines were mostly trash.

You needed to search all of them to find something decent.

That's roughly analogous to today. Ignoring cost, you'd be way better off asking all the LLMs to solve a problem (like coding) where you can verify the answer.

So the question is, for things like that -> can a group of models perform better than frontier models, especially at a reasonable cost?

Fable is not a great value, so unless you're trying to find answers to Erdos questions, you can probably do better on cost.

You can probably typically ask 3 or 4 of the top Chinese models for an answer and get a response for the same Fable question... Given that Fable isn't that much better, it's not surprising you can do better for a large subset of problems.

Re:Show HN: Echo – Fable-level results at 1/3 the cost using... by Terretta on Thursday July 23, 2026 @07:04PM (#49029250)
> Dogpile was only a good idea while Search Engines were mostly trash.

Precisely.

Re:Show HN: Echo – Fable-level results at 1/3 the cost using... by user_7832 on Friday July 24, 2026 @12:05AM (#49031109)
LLMs fail in such bizzare, obscure methods (to the average observer) at times. Sometimes even simple questions ("who was that x person who was super famous I'm thinking of") type questions fail terribly.

The more vague and non committal and hand-wavey and subjective the field for AI to answer, the better the results (imo).

Re:Show HN: Echo – Fable-level results at 1/3 the cost using... by rablackburn on Friday July 24, 2026 @12:18AM (#49031185)
I daresay it's because the assistant training corpus is heavily biased towards one-shot solution answers.

Because the correct response to that query is "I have no idea -- you will need to provide more information"

and LLM Agents suck at that.

Re:Show HN: Echo – Fable-level results at 1/3 the cost using... by ai-x on Thursday July 23, 2026 @10:22PM (#49030572)
There is no reason a smarter model will not build an internal smarter/efficient router itself
Re:Show HN: Echo – Fable-level results at 1/3 the cost using... by joe_the_user on Friday July 24, 2026 @12:27AM (#49031218)
Dogpile was only a good idea while Search Engines were mostly trash.

Well, search engines are trash again. Perhaps it should come back

Re:Show HN: Echo – Fable-level results at 1/3 the cost using... by onlyrealcuzzo on Friday July 24, 2026 @08:11AM (#49034430)
> Well, search engines are trash again.

That's mostly because the ability to make money on the web made the web trash.

When you couldn't monetize your websites, everyone's websites were passion projects.

Now everyone is trying to figure out how they can make $8M a year off yet another recipe website.

Re:Show HN: Echo – Fable-level results at 1/3 the cost using... by tiffanyh on Thursday July 23, 2026 @06:35PM (#49028968)
Exactly

High quality, fast & cheap (all 3 combined) - is a formula success.

It’s just way easier said than done.

Re:Show HN: Echo – Fable-level results at 1/3 the cost using... by Avicebron on Thursday July 23, 2026 @06:44PM (#49029064)
There's a reason most pros will tell you pick two of the three.
Re:Show HN: Echo – Fable-level results at 1/3 the cost using... by alizaki on Thursday July 23, 2026 @05:27PM (#49028283)
Indeed. What a deep cut
Re:Show HN: Echo – Fable-level results at 1/3 the cost using... by ignoramous on Thursday July 23, 2026 @05:42PM (#49028446)
Mixture of Models, perhaps? tbf to OP, the setup they're proposing has also been recently evangelized by other "ai gateway" products (like OpenRouter, JusCode, Fireworks etc), so there's likely something useful here.
Re:Show HN: Echo – Fable-level results at 1/3 the cost using... by fmajid on Thursday July 23, 2026 @06:07PM (#49028700)
Re:Show HN: Echo – Fable-level results at 1/3 the cost using... by johnvanommen on Thursday July 23, 2026 @07:29PM (#49029480)
This sounds like networking.

Did OP invent an “intelligence router?”

Re:Show HN: Echo – Fable-level results at 1/3 the cost using... by cdelsolar on Thursday July 23, 2026 @05:28PM (#49028299)
timecube?
Re:Show HN: Echo – Fable-level results at 1/3 the cost using... by ljlolel on Thursday July 23, 2026 @08:50PM (#49030027)
ensemble models always did the best at Kaggle

we did the same: https://trustedrouter.com/blog/prometheus-2-new-draco-state-...

Re:Show HN: Echo – Fable-level results at 1/3 the cost using... by ignoramous on Friday July 24, 2026 @12:25AM (#49031210)
You need update your blog posts.

  Our whole stack is radically open source — frontend and backend alike, Apache-2.0 licensed — and so is everything behind this benchmark. That is how a benchmark number earns trust: verifiability, not hype.
The repos have since been moved to BUSL-1.1: https://github.com/Lore-Hex/quill-router/commit/8155ac666ae0...
Re:Show HN: Echo – Fable-level results at 1/3 the cost using... by ljlolel on Monday July 27, 2026 @10:49AM (#49070506)
everything has been updated
Re:Show HN: Echo – Fable-level results at 1/3 the cost using... by moralestapia on Thursday July 23, 2026 @09:33PM (#49030285)
metacrawler.com
Re:Show HN: Echo – Fable-level results at 1/3 the cost using... by gchamonlive on Thursday July 23, 2026 @09:46PM (#49030364)
Well, you can try to make triangular wheels, but they are round for a reason
Re:Show HN: Echo – Fable-level results at 1/3 the cost using... by hahahaa on Friday July 24, 2026 @01:48AM (#49031632)
This is the don't use the same EC2 size for everything on your service approach.