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.
agreed, unless you need to use all models i'm sitting here wondering why orgs would want to introduce third party risk into their pipelines for marginal cost and time savings
I've been off-and-on again. I stopped using it for years. I begrudgingly began using it again as I founded a private game server (before AI was viable, even, so it wasn't sued) for a game that shutdown a few years ago.
Perhaps hilariously, the moderators there also didn't like me and told me to eat shit when I suggested people start doing packet captures. At least they warmed up to me over the years though and actually endorse the project now.
Doesn't make the rest of the website any less of a cesspool though.
No OP, but MCP really is just a logical next step once you've got an API.
The API is the "low level" protocol, the MCP is the high level one, suited perfectly to an LLM that can call tools (since MCP essentially turns an API into a LLM tool).
With just an API, the agent needs to "read your API docs" to know how to call it (that can be an OpenAPI spec or even just text).
With MCP, the agent sees a bunch of tools it can call, and they've been trained to call tools so they nail it.
One more very important factor is authorization, which no one seems to mention in these discussions. CLIs were made for humans and use primitive mechanisms for authorization: either an API key you hardcode in your environment, or they literally run a background HTTP Server to get a callback OAuth call to receive a token from a browser authentication flow. Incredible that people are happy with that, appparently. With the MCP Authorization spec, you solve authorization across multiple MCP servers in the same standardized way, the LLM client you use just need to know the protocol, not how to login for every single MCP server.
Very importantly, if the MCP client does the authorization, the MCP provider has auditability: is this a call from a human or from a LLM? That's important in Enterprise! People think it's ok to let an LLM act on behalf of the human but that will eventually bite a lot of people. Did the LLM just try to hack the API while you were mindlessly clicking "yes" when it asked if you wanted to let it do something? Tough luck, there's no way to distinguish an LLM making a mistake from a human maliciously running some attack.
And as the post mentions, there's also more benefits like being able to "elicit" user input (not just request/response cycles) and the ability to have documentation and assets (skills also have this though).
This is a great example of the AI-hype-induced reply.
> to an LLM that can call tools (since MCP essentially turns an API into a LLM tool).
"Tools" is literally an API call
> With MCP, the agent sees a bunch of tools it can call,
Yes, the agent first calls a specific API that returns the schema for that particular server. It's literally the same.
> One more very important factor is authorization, which no one seems to mention in these discussions.
Yes, API calls to services are often gated behind auth. OAuth that MCP uses is from 2006, and its version 2 is from 2012. What do you think it was created for?
> the MCP provider has auditability: is this a call from a human or from a LLM? That's important in Enterprise
We had "differentiate these two accounts and audit log their activity" probably since the 1950s
> there's also more benefits like being able to "elicit" user input
Two-way communication is also a thing since the 1950s, probably.
If you think tool call and letting the LLM call an API via curl are the same thing, you haven’t a clue how LLMs work and honestly shouldn’t be commenting on the topic at all.
Wasn't even aware Mistral was around and I think that just shows you how irrelevant it has become and not a very good sign for EU in general when the best talent are working for American AI companies.
I saw Tibo's tweet a while back and it was basically a legitimate complaint about the extreme taxation he faced back in EU (France I think) and its pretty obvious how much of a hinderance top down centralized regulation is to innovation.
While I welcome competition and independence, nobody can argue with American innovation and its ability to attract the best of the best. Once it takes seat of the AI reigns there is very little chance for other countries to compete, very much similar to semiconductor field and how only a few select countries have the talent and monopoly over its particular supply chain.
It's clear to anyone looking in that whatever EU is doing is not working (not just AI) and will not work as they do not seem flexible or humble enough to steer itself.
Big tech has remote offices in every major European economy, and they pay well above top 90th percentile of market rate. It basically has a talent sucking effect on the entire economy.
You are most welcome and thank you for letting me know you like it!
I have seen the buffer and repeat issue with some channels. It's hard to know why it's happening. The app works by giving the stream URL from the broadcaster to a video player element in your browser. It just uses standard web protocols and tries to get the data from the stream as fast as it can. I've found these international sources can be iffy sometimes. Others are rock solid and you can watch high dev TV like it's from your home town!
https://youngit.blogspot.com/2012/08/ufo-recorded-in-annals-...
reply