It took me quite a while to come round to OpenRouter. Originally I didn't understand why anyone would put a proxy between them and an LLM, but it actually adds some quite significant value:

1. By far the lowest friction way to support and try out all the models.

2. They offer billing caps! Most model providers still don't do this [EDIT: maybe they do, see reply comment], but if you're going to run anything in public it's very useful to have hard limits so it doesn't cost you $1m overnight because someone started abusing it.

3. Their rankings are one of the more interesting signals for which models are popular, despite their flaws (most OpenAI and Anthropic users don't go via OpenRouter, it's currently not possible to tell the difference between many users switching v.s. one "whale" changing their preferred model)

Given how API costs are becoming meaningful for a lot of companies now, having a provider like OpenRouter to help measure your spend and easily experiment with and switch providers feels like a valuable service.

As someone who uses OpenRouter extensively (and wrote an unintentional adjacent PR piece a few days ago: https://news.ycombinator.com/item?id=48317294 ), it's definitely the best way to try out new models without fiddling with each providers distinct APIs which is becoming a recurring concern as of late.

That said, I don't understand the people who use something a full agentic backbone with expensive models like Claude Opus with OpenRouter because that 5% surcharge is meaningful at that level of cost instead of going with the source API providers. But people are clearly doing it, and it's pure revenue.

U can broadcast your tokens I wonder how to do something with such data. Aside from greping for secrets

https://openrouter.ai/docs/guides/features/broadcast

One thing that OpenRouter makes easy is the ability to manage API keys (mint new ones, expiry/limits per key, etc.) that I wish that other providers would make possible/easier.

So many use cases, like sharing AI/assisted features externally, with the ability to use those features but also limit the fallout if its shared / used for other purposes, without jumping through more fallible hoops like safeguards etc.

Is the Open in OpenRouter the same as in Open AI? I couldn’t find any repository or hosted code. Thought it'd be a open source, self hostable tool with a cloud offering but seems its just the latter?

I think that OpenRouter will continue to be very popular while there lots of experimentation in the LLM space, and while the "current favorite" model continues to change between various frontier labs.

After things begin to settle down, we'll probably see a consolidation of both frontier and open-source models - and then OpenRouter will become less useful, because that 5% overhead is well worth it when you want to try 20 models from 10 labs, but harder to stomach when you only need 5 models from 2 providers, and each of those providers has its own API knobs that you can tune to make things even cheaper.

I was sort of hoping that they were bootstrapped or at least non-VC funded. I'm wary of them introducing consumer-unfriendly revenue-generating schemes.

At what valuation?

I’m a user and I like the routing layer and not having to change things up too much, but I’m not sure why a solid business model for this product would require this much money at this kind of valuation unless they’re trying to buy data center capacity to self-host models eventually?

The moat will disappear in 1-2 years

Congrats to the OpenRouter team for securing this round of funding. The 5% surcharge for their pricing model may not be palatable to enterprises. In fact, the OpenRouter team could be a pivotal part of the enterprise GenAI stack if they can allow configurable, pluggable endpoints for routing directly to enterprise vetted endpoints to 1P/3P LLM APIs. A couple of large companies I’ve worked so far kinda have this system in place, albeit the dev and maintenance cost and of setting up such an “LLM gateway” could be significantly reduced with OpenRouter. I feel that this is largely an ignored, forgotten part of operating GenAI apps at scale.

One thing I haven't seen mentioned here yet and really like about OpenRouter is their openrouter "meta" model, that automatically routes the prompt to an appropriately capable model. Saves me a ton of money on not routing everything through Opus, but not giving me bad results when I ask something more complex, which gets autorouted to Opus.

Ok I’ve read the comments I still don’t see it. What’s the 113M in value add here?

Using Tinfoil, Replicate, Cerebras, and OpenRouter. Competition is good.

How quickly they get new models supported on the API and it just works, is insane!

I still don't get the value proposition: You rarely have to use all the models, you will likely end up with a few for your workflow but there is a way to use them/try all if you wanted to, neato.

Also one scary issue I had with OpenRouter in the early days, I think I saw somebody else's context and there were weird Chinese characters, haven't touched it since.

What's the business model? Their core functionality, while useful, seems like something that will just be an open-source package. I assume there will be some Saas layer on top of it?

"Over the last six months, weekly volume on OpenRouter has grown from 5 trillion to 25 trillion tokens".

DAMN!!

That's 41+ million tokens every second. That scale is crazy for such a small team of 48-50 people overall.

OpenRouter’s biggest value to me is reducing switching costs between models. The markup matters at scale, but for exploration and early-stage development, the convenience is hard to beat.

OpenRouter is our primary provider for evaluation data, and we've been really happy with them!

I'm sure they're experiencing growing pains, but a larger model selection (and faster releases for open weights models), would keep us from using other providers. For example, it took much longer than it should have to get Qwen 3.6 ~30B class models released (almost 2 weeks if I recall)

undefined

You could vibe code something like this in probably under a month and there’s zero moat here. Clown economy. And I like and use OpenRouter.

An amazing service. I use its 20+ free LLM options to allow completely free usage of LibreOffice AI extension with no signup https://librethinker.com .

Too bad api use is like 100x more expensive than subscriptions for the big 3.

Why does a company with a seemingly health business model that is already churning profits and doesnt require large CapEx, taking losses to capture users, need to be raising this kind of capital?

> ... with participation from NVentures (NVIDIA's venture capital arm), ServiceNow Ventures, MongoDB Ventures, Snowflake Ventures, Databricks Ventures ...

Are tech companies FOMOing so hard that they're now all running AI venture arms themselves instead of you know, developing their own products? Except for NVIDIA who needs to keep pumping the bubble I didn't expect the others.

Every tech company now seems to be invested in every AI startup.

I honestly thought this was some kind of OpenWRT firmware for routers until I clicked the link. "Ahhh, AI. Of course."

Aren’t they totally dependent on the good will of the model providers?

I like OpenRouter - lets me test out new model quickly and easily. I would still need a good functioning mobile application for it.

I think they should go in this direction: they should make their own Model Agnostic versions of whatever functionalities other AI companies are making. Examples

1. personal chat app

2. the chat app working with their own implementation of memory

3. coding harnesses that are model agnostic

When I think of OpenRouter, I should think of "model agnostic LLM tools".

undefined

[flagged]

[dead]

[dead]

[flagged]

[dead]

I’m offering a fully end to end encrypted open source version and hosted version of open router : https://trustedrouter.com/