“Current routing options were charging 5% fees for byok plus provider fees (openrouter) or were built in heavy python packages with ecosystem restraints. I wanted to be completely free in terms of use, and not have to pay extra for tokens.”
“I use SillyTavern mobile using OpenRouter to get providers to hook to it to try to roleplay. Before I used to use Deepseek v4 (now 0423) with Baidu Qianfan provider and it worked well 85% of the time, meaning no refusals.”
“Openrouter tracks what apps are using the model and the top ones for hy4 are all different coding harnesses.I guess it could be fake but seems more likely people are just trying it out.”
“That's not to say that specific providers might have issues or worse cache implementations - but it stands that if openrouter is juggling the requests back and forth by default then _that alone_ is breaking caches on those requests in huge numbers.”
“currenly I use pi with openrouter and is very nice I combine with a chatgpt sub that allows me to use it with pi. and when my usage runs out I turn to openrouter. Then same when I need anthropic models”
“I finally configured claude desktop app with openrouter as a gateway, and got composio connected so claude can access 500+ web apps. However, I get this error when trying to run anything and I'm stuck:”
“if i had exactly $10 and needed reliability, id skip the free tier entirely and just use a cheap paid model. llama 3.1 8b on openrouter is something like 2 cents per million tokens.”
“openrouter data is super interesting for tracking real api demand shifting from proprietary to open weights. usually see spikes when a new fast model drops, but production traffic takes longer to switch due to prompt tuning and custom function calling setups.”
“What I take from it: the free stealth routes are real enough to ship small, well-specified features through a full gate stack, and the caps plus retention plus anonymity mean they are for code you'd happily paste in public.”
“Yea, that’s pretty normal on OpenRouter. Since they have a huge number of free users, the moment a new model becomes available for free, a ton of people jump on it and the traffic spikes. If you mean the rate limit will go down, I’m not sure.”
“I keep trying to use z-ai/glm-5.2:free on openrouter since GLM 5.2 is perfect for me but too expensive for my usage, but I keep getting rate limited even though im literally not even using it, Ill try chatting with it directly on openrouter and not a single message goes through.”
“Laws apply here. If OpenRouter/OpenCode are lying then there is potential legal risk for them. If a Chinese provider is lying, it would be harder, but there is a reputation risk at least.”
“OpenRouter is really strong in developer-type tools where you want a simple way to pick a model. It's really strong with chatbots where they don't have to be perfect.”
“I guess we will see, the good news is that they don't offer anything really that special, if they mess up and people want alternative, it's not that hard to make one, there are even a few router that works like openrouter”
“OpenRouter is adopting the Netflix strategy. If you're just serving everyone else's catalogue, there's no friction for switching costs to Amazon Prime, etc.”
“Startup teams with limited time that are focusing their energy on finding the right product to build also use openrouter.I think the bet is that you offer enough value-adds that companies who find traction don’t want to switch to something enterprise grade, and you grow into being enterprise grade with them.Very similar to Stripe’s trajectory, incidentally.”
“Which can deal with a safe-for-work beach photo without a censorship system false alarm?OpenRouter lets me run my tests against openai and anthropic and google and x and bytedance and qwen and llama, with a single sign-up and a single payment.”