frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

MicroVMs: Run isolated sandboxes with full lifecycle control

https://aws.amazon.com/blogs/aws/run-isolated-sandboxes-with-full-lifecycle-control-aws-lambda-in...
137•justincormack•3d ago•77 comments

Show HN: Smart model routing directly in Claude, Codex and Cursor

https://github.com/workweave/router
57•adchurch•1h ago•40 comments

Ultrasound imaging of the brain

https://alephneuro.com/blog/ultrasound-brain
143•rossant•6h ago•53 comments

Previewing GPT‑5.6 Sol: a next-generation model

https://openai.com/index/previewing-gpt-5-6-sol/
313•minimaxir•1h ago•296 comments

Om Malik has died

https://om.co/2026/06/24/1966-2026/
1206•minimaxir•22h ago•143 comments

An entire Herculaneum scroll has been read for the first time

https://scrollprize.org/firstscroll
1570•verditelabs•1d ago•344 comments

Modern GPU Programming for MLSys

https://mlc.ai/modern-gpu-programming-for-mlsys/
8•crowwork•3d ago•1 comments

Springer Nature has removed two studies by Max Planck

https://www.science.org/content/article/why-have-papers-one-history-s-most-famous-physicists-been...
268•adharmad•4h ago•122 comments

Liva AI (YC S25) Is Hiring

https://www.ycombinator.com/companies/liva-ai/jobs/gvtc3Ep-founding-operations-lead
1•ashlleymo•1h ago

A US military exercise to launch a satellite on short notice

https://arstechnica.com/space/2026/06/a-us-military-exercise-in-space-got-underway-with-barely-an...
88•jonbaer•3d ago•49 comments

Libre Barcode Project

https://graphicore.github.io/librebarcode/
257•luu•15h ago•43 comments

Incident CVE-2026-LGTM

https://nesbitt.io/2026/06/26/incident-report-cve-2026-lgtm.html
406•mooreds•5h ago•73 comments

Mullvad founder Daniel Berntsson donated millions to far right party

https://www.na.se/orebro/miljonregn-till-orebropartiet-fran-techprofil/
7•mrtesthah•1h ago•0 comments

LaTeX.wasm: LaTeX Engines in Browsers

https://www.swiftlatex.com/
41•theanonymousone•2d ago•11 comments

22-year-old Mozart's handwritten notebook unearthed in 'major discovery'

https://www.classicfm.com/composers/mozart/handwritten-notebook-discovered-major-paris/
217•thunderbong•6d ago•65 comments

What happened after 2k people tried to hack my AI assistant

https://www.fernandoi.cl/posts/hackmyclaw/
315•cuchoi•16h ago•145 comments

What is a Lithium-ion capacitor?

https://www.jtekt.co.jp/e/products/capacitor/capacitor_about.html
9•ksec•2h ago•1 comments

Framework's 10G Ethernet module exposes USB-C's complexity

https://www.jeffgeerling.com/blog/2026/framework-10g-ethernet-module-usb-c-complexity/
287•Alupis•17h ago•165 comments

What Is a Nomogram and Why Would It Interest Me?

https://lefakkomies.github.io/pynomo-doc/introduction/introduction.html#what-is-a-nomogram-and-wh...
3•Eridanus2•1h ago•0 comments

Bipartite Matching Is in NC

https://scottaaronson.blog/?p=9851
91•amichail•3d ago•13 comments

'Cost Me the Election': Data Centers Trigger Voter Backlash

https://www.newsweek.com/cost-me-the-election-data-centers-trigger-voter-backlash-12118327
29•randycupertino•1h ago•23 comments

Show HN: WebBase-III – dBASE III rebuilt in the browser with its own interpreter

https://github.com/DDecoene/WebBaseIII
58•ddecoene•2d ago•15 comments

Jolla Phone (October 2026)

https://commerce.jolla.com/products/jolla-phone-october-2026
189•mrbn100ful•3h ago•114 comments

A game where you're an OS and have to manage processes, memory and I/O events

https://github.com/plbrault/youre-the-os
330•exploraz•3d ago•71 comments

The 'papers, please' era of the internet will decimate your privacy

https://expression.fire.org/p/the-papers-please-era-of-the-internet
1018•bilsbie•20h ago•516 comments

The Garbage Collection Handbook: The Art of Automatic Memory Management (2nd Ed) (2023)

https://gchandbook.org/
221•teleforce•19h ago•46 comments

U.S. government will decide who gets to use latest upgrade to ChatGPT

https://www.washingtonpost.com/technology/2026/06/26/openai-says-us-government-will-vet-users-its...
5•alain94040•12m ago•1 comments

Ask HN: Is "no source code was copied" still a sufficient copyright defense?

18•oscgam1•3h ago•18 comments

FEXPRs vs. vtable: how LispE interpreter works

https://github.com/naver/lispe/wiki/2.7-FEXPR-vs.-vtable
68•birdculture•2d ago•6 comments

We all depend on open source. We will defend it together

https://akrites.org/letter/
411•dhruv3006•12h ago•198 comments
Open in hackernews

Show HN: Smart model routing directly in Claude, Codex and Cursor

https://github.com/workweave/router
56•adchurch•1h ago
We built a model router that plugs into coding agents (e.g. Claude Code, Codex, Cursor, etc.) and intelligently sends requests to the best model to serve them. Here's a quick demo of running it locally: https://www.youtube.com/watch?v=isKhAyivtfM.

At Weave, we write most of our code with AI, and it's been getting more expensive. This came to a head when Opus 4.7 was released and, thanks to its tokenizer changes, our costs shot up. We knew we didn't need Opus for everything but we didn't want to lose out on the intelligence for the cases where you really need it. So we decided to build a model router to handle this for us.

The Weave Router acts as an Anthropic/OpenAI endpoint specifically for coding agents. It looks at every inference request and intelligently (more on that in a sec) decides what model to send it to, handling all the translations required along the way. So it can use faster/cheaper models (e.g. DeepSeek v4, GLM 5.2, Kimi K2.6) when possible, and frontier models (Opus 4.8 & GPT 5.5 (& Fable whenever it's back)) when necessary.

How do we know what model to route to? We trained an RL model on tens of thousands (so far!) of agent traces. We reward the routing model when it selects an LLM that successfully completes the given task.

Here's an example: if you ask the router to plan a complex change, it will (probably) route that request to Opus 4.8. Subagents exploring the codebase to gather context will be routed to more suitable models (e.g. DeepSeek V4 Flash). Then when you have the plan ready to implement, it will be (most likely) be handed to a quicker model (e.g. GLM 5.2) to carry it out.

We've been using this internally for the last month or so. We've saved 40% on tokens vs. what we otherwise would have paid, with no noticeable differences in quality or velocity.

The router is source-available under Elastic License 2.0, so you can self-host it. Or if you prefer, you can also use our hosted version: weaverouter.com.

I'll be here to answer any questions you may have!

Comments

stpedgwdgfhgdd•1h ago
The thing I do not get with these routers is that you will have more cache misses (5min ttl). And if there is one thing i’ve learned; using the cache is crucial.

How does this router translate to $$$ when developing?

adchurch•1h ago
You're right and that's why we built the router to be cache aware! Once it starts using one model, the threshold to switch to another model will be higher because the additional cost of the cache miss needs to be worth the cost savings or quality increase.

This is the key thing that other routers we've seen miss: they're stateless so for a coding agent use case you end up spending more money due to all the cache misses.

alansaber•47m ago
That is interesting, sounds like in practice you only end up routing between 2 models
adchurch•45m ago
I'd say that a typical main agent loop has 1-3 models (obviously very situationally dependent), but when you have subagents those can get routed independently since they have a fresh context window, so there are a lot more degrees of freedom there.
echelon•14m ago
Or not routing at all.

In practice you just pick one and stick with it until the API stops or you hit performance issues.

adchurch•7m ago
The choice on the first turn is super important for this reason! But if a user prompt sends the convo in a very different direction then often it does make sense to reroute at that point.
_pdp_•1h ago
Cool.. but I still don't get how this is going to save money. It seems to me that it might actually burn more money just because the whole system now seems to be coming from different LLMs.

Also, small LLMs are prone to stop before completion, throw errors and produce loops. Is this factored in the design of the tool? I am not sure.

edit: spellcheck

adchurch•58m ago
It saves money because some agent sessions can be entirely handled by a smaller model (also relevant: subagents use fresh context windows so a subagent with a simple task can be routed to a smaller model even if the main agent needs a frontier model).

Totally right about small LLMs btw, that's why we trained this on real agent sessions where we forced it to use different models. If the routing model sees small models can't handle a certain type of task then they won't be assigned. (Also as a fallback we have some guardrails that will have a bigger model come in to "rescue" a smaller model if it gets stuck)

arendtio•1h ago
What is the difference from Cursors 'auto' mode?
adchurch•56m ago
Fun fact: Cursor's "auto" mode is just Composer (or at least it was last time I checked). So it's different in the sense that it actually does route to more than 1 model
ai_slop_hater•1h ago
Isn't this more expensive than always using the same model, since, as I understand, by routing to different models you give up on cache?
adchurch•53m ago
If you statelessly route each new request: yes it does end up being more expensive!

So our routing is cache-aware. It will have a much higher threshold to switch from one model to another if there's already some cache for the first model. Experimentally this solves the problem (like I said we've saved 40% ourselves vs. what we would have otherwise paid).

debarshri•58m ago
It is funny. We are building something similar.
adchurch•51m ago
Oh cool, feel free to reach out to me at andrew@workweave.ai if you ever want to share notes! We've learned a lot in the process of building this so far :)
spqw•57m ago
This + making sure common requests are saved as reusable skills and scripts would probably save a large part of my token usage

As prices increase we will see more of these tools to optimise and make the best use of token budget

adchurch•43m ago
100%, from what we've seen, for a lot of big companies that 1. don't have subsidized usage and 2. are pushing AI adoption hard, figuring out token costs is P0 or P1 for their eng leadership
SoftTalker•12m ago
So you're saying that since adopting AI/LLM tech many companies have their top engineering priority being optimizing the costs of that rather than ... addressing actual business needs?
g00k•53m ago
Man, I'm not so sure if I'd use something like this because the way I prompt already changes based upon what model I am using. I'm not convinced it would route to the right model based on my diction or whatever.
alansaber•47m ago
Yep this was always the reason to avoid "auto" mode in cursor.
adchurch•47m ago
Yeah that's a really interesting point, tbh I think the more relevant variable here is the harness you're using rather than the specific model? i.e. GPT 5.5 in the Claude harness behaves a lot more like Claude than Codex if that makes sense.

Hard to quantify this ofc but that's what I've felt vibes wise from using this for the last month.

emilio_srg2•51m ago
but this means you work with API pricing rather than subscription pricing. Isn’t it better to use claude or codex CLI etc directly in terms of cost?
adchurch•49m ago
If you have a Claude/Codex subscription then we use that (and account for the subsidized price accordingly when making routing decisions) instead of API billing. So you get the best of both worlds: subsidized usage for frontier models + save by using open/smaller models when it's genuinely better.

In practice, lots of ppl are using this to make their Claude sub limits go further!

emilio_srg2•38m ago
I see but didn’t they severely limited the usage allowed with `claude -p`
adchurch•36m ago
But we're not routing via `claude -p`, if you have sub usage available + it's the right choice to route to a Claude model, then the router is approximately a transparent passthrough. So it gets billed like normal `claude` usage rather than `claude -p`.
alansaber•49m ago
"We reward the routing model when it selects an LLM that achieves the task successfully" sounds pretty oversimplified
adchurch•33m ago
Indeed it is :) I skipped over talking about all the RL machinery, network design, reward function design, state representations, etc. because really the intuition is that we tell the model when it accomplishes its goal, and then it learns over time how to get better at making the right decisions in order to accomplish its goal.

Happy to talk about this in some more depth if there's anything specific you're curious about!

gautam_io•46m ago
This is cool!

Will this use my Claude Pro/Max subscription? Or will it always use the API billing "pay as you go"?

adchurch•45m ago
Yep it uses the Claude sub if possible and falls back to API billing only if you don't have a Claude sub or it's out of usage! Same deal for Codex
slopinthebag•43m ago
> At Weave, we write ~all our code with AI

This is probably not a very effective way of marketing imo. At least, it turns me completely off.

suyash•42m ago
I would rather just use OpenCode - leverage AI models, even can host locally or paid ones with ease.
adchurch•37m ago
We integrate with OpenCode too! OpenCode provides the harness, then the router selects the right model for the task.

We haven't yet set up local model routing though, that's really interesting - have you had any success using local models for coding tasks? Tbh I haven't heard many success stories from using local models yet

k9294•37m ago
What about request caching? If you swap to a cheaper model mid execution it might cost more that to make multiple requests to the already cached provider?
adchurch•31m ago
Yep 100%, mentioned this in another thread (https://news.ycombinator.com/item?id=48689448) but tl;dr we build the router to be cache aware
mkagenius•30m ago
We have created Murmur[1] which kind of works with your existing subscription (having API key is not mandatory). You can just tag @copilot @codex from claude code to delegate work to them. (it can also do it on its own too btw)

1. https://github.com/instavm/murmur - Murmur

adchurch•18m ago
Very interesting - curious how you've used it yourself so far? I can imagine one use case would be having e.g. GPT 5.5 review Opus 4.8's work?
iluvcommunism•29m ago
This is basically what I need, a router. I’m tired of changing intelligence & speed levels manually.
adchurch•17m ago
Nice, let me know any feedback you have from trying it out!
peterbell_nyc•21m ago
I auto tune my prompts to a locked model version based on production data used as evals with holdback data. I think the use case for this would be one off interactive prompts? For now I just run those all against an Opus 4.8 MAX and I'm sure I could downtune, although for interactive my opening prompt isn't always reflective of my overall goals for the multi turn session.

I'm just trying to figure out why on the fly routing would beat testing and tuning and locking models and versions for each class of call, with evals and auto tunes running to explore more possible models for commonly run classes of prompt over time . . .

bijowo1676•6m ago
How come data privacy and confidentiality is not an issue with services like these?

Do people voluntarily use these proxies/routers, knowing their prompts, outputs and code will be seen by other people ?

I get it might be ok for personal projects, but for anything that makes money and is a part of business... this must be big no-no ?

jakozaur•4m ago
It's rather hard to do at the proxy level with agentic coding, such as Claude Code or similar. These are long-chained sessions of tool use that heavily rely on prompt caching. Changing mid-flight is costly.

It looks like much more context is required to decide on the best model (e.g., summarizing logs might use a cheap model, whereas you likely want Opus/Mythos/GPT 5.6 to debug multithreading logic). In an agentic system, a decision about the model may be embedded in the decision to orchestrate the model.