# OpenRouter 666 Hacker News commenters recommended OpenRouter and 321 criticized it, November 2025 to November 2026. It was recommended in 13 of 13 months; owners who said how long report a median of 0.5 years (8 owners). It meets the Hacker Approved badge rule. **Verdict:** One API and credit balance for nearly every AI model, making it easy to switch models, pick cheap providers and cap spending; critics flag quantized providers and broken caching. ## Why people recommend it - Easy switching between many models (7 comments): "the true advantage of OpenRouter is that, in a space where there's a new model being released every week, you can easily switch to whatever model is best at any given time" ([source](https://news.ycombinator.com/item?id=49365705)) "You can easily switch models, use the cheapest provider (especially for open models), and not have to reach certain "tiers" to get access to limits" ([source](https://news.ycombinator.com/item?id=46209599)) "Rather than building API clients for five different AI providers, you build one client to OpenRouter, and switching models become extremely easy." ([source](https://news.ycombinator.com/item?id=49323831)) - Single account, key and billing (4 comments): "it's easier to go through one provider (OpenRouter) instead of having the hassle of managing different endpoints and accounts" ([source](https://news.ycombinator.com/item?id=47360435)) "it's an excellent way to simplify the experience of hopping between model providers with a single unified billing system" ([source](https://news.ycombinator.com/item?id=49661226)) "The big advantage of OpenRouter IMO is that you don't have to manage keys, URLs, and subscriptions with 10+ providers." ([source](https://news.ycombinator.com/item?id=49665889)) - Spend limits and policy controls (4 comments): "since being able to charge 5% on your AI spend just for having sane billing, spending controls (actually enforced per-api-token budget limits!) and easy sign-up" ([source](https://news.ycombinator.com/item?id=46231847)) "OpenRouter's value proposition is less about avoiding two hour developer tasks and more about having a single place to establish policy controls" ([source](https://news.ycombinator.com/item?id=49665458)) "The main thing about Openrouter is also that they take 100% of the risk in case of overcharges from the models, you have an actual hard cap." ([source](https://news.ycombinator.com/item?id=47705462)) - Cheap access to open models (4 comments): "You can easily switch models, use the cheapest provider (especially for open models), and not have to reach certain "tiers" to get access to limits" ([source](https://news.ycombinator.com/item?id=46209599)) "I just connect to Openrouter and use something like Kimi, GLM or Deepseek for a fraction of the price of Anthropic et al." ([source](https://news.ycombinator.com/item?id=48737725)) "that would've cost way more than double or quadruple per token at the big 3 if I didn't have those alternative options" ([source](https://news.ycombinator.com/item?id=49775603)) ## What people warn about - Provider switching breaks prompt caching (4 comments): "OpenRouter isn’t capable of caching as effectively as DeepSeek is because they will, for instance, switch inference providers in the middle of a session" ([source](https://news.ycombinator.com/item?id=49899647)) "Without pinning providers, you pay let’s say 2x more for agentic coding, since that many input tokens aren’t cached as you bounce between providers." ([source](https://news.ycombinator.com/item?id=49657245)) "if openrouter is juggling the requests back and forth by default then _that alone_ is breaking caches on those requests in huge numbers" ([source](https://news.ycombinator.com/item?id=49494352)) - Cheap providers serve degraded quantized models (3 comments): "This is even worse via OpenRouter, with providers serving open-source models, some of them serve heavily quantized versions or even completely different models." ([source](https://news.ycombinator.com/item?id=48548412)) "Providers like OpenRouter default to the cheapest provider. They are often cheap because they are rediculously quantized and tuned for throughput, not quality." ([source](https://news.ycombinator.com/item?id=47842984)) "I don't want to control routing. I want the model to work how the giant, prominent "BENCHMARKS" section says it works, not randomly have a 100x error rate." ([source](https://news.ycombinator.com/item?id=49657212)) - Extra middleman sees your data (3 comments): "I assume they (or their downstream providers) are MITMing all my requests and are thus unsuitable for any professional use." ([source](https://news.ycombinator.com/item?id=48341560)) "Because you have Cloudflare (MITM 1), Openrouter (MITM 2) and finally the "AI" provider who can all read, store, analyze and resell your queries." ([source](https://news.ycombinator.com/item?id=46264298)) "Openrouter yes sees everything. But as long as they are independent and establishing their core business on privacy and confidentiality reputation, I feel more comfortable with this than with a new broker." ([source](https://news.ycombinator.com/item?id=49326330)) - Added layer hurts reliability (3 comments): "Adding another layer on top of Google's own APIs adds latency, lowers reliability, and (AFAIK) doesn't allow batch mode" ([source](https://news.ycombinator.com/item?id=46226151)) "I've not tried OpenRouter as adding yet another layer will just make it that much more difficult to get a model and provider combination working well." ([source](https://news.ycombinator.com/item?id=49662373)) "insane that openrouter's main feature is literally a fallback and streaming doesn't support 200 no content to fallback to another provider or smth." ([source](https://news.ycombinator.com/item?id=49657000)) ## Alternatives hackers recommend - [OpenCode Zen](https://hackerapproved.com/p/opencode-zen.md): 36 recommend - [Cortecs](https://hackerapproved.com/p/cortecs-ai.md): 9 recommend - [Cloudflare AI Gateway](https://hackerapproved.com/p/cloudflare-ai-gateway.md): 9 recommend - [Eden AI](https://hackerapproved.com/p/eden-ai.md): 9 recommend - [TrustedRouter](https://hackerapproved.com/p/trustedrouter.md): 8 recommend ## Where to get it - [Visit the official site](https://openrouter.ai/) ## Cite as Hacker Approved, "OpenRouter", https://hackerapproved.com/p/openrouter/ (counts from 1,795,640 Hacker News comments, November 2025 to November 2026).