How to keep AI spend flat while token usage grows exponentially: Not with friction and spend alerts. With better defaults, routing, and caching.
Better Defaults (not Usage Caps) – Engineers can choose any model they want, but defaults matter. We’re experimenting with defaulting to open weight models like GLM 5.2 and Kimi 2.7 through our LLM gateway, while still encouraging engineers to choose the right model for the task. 91% of our employees were never hitting their usage caps, so instead of lowering caps and driving up alerts, we're moving to cheaper defaults. Note that code reviews use a diversity of models, so they can check each other's work.
Better Routing – In our custom harnesses, we preprocess prompts and route to the best model for the job, considering cache hits and model pricing. For instance, you may want a frontier model for planning, but not for execution where they can be overkill. Ultimately, humans shouldn't be choosing models - AI can automate this task.
Better Caching – Cache misses are the easiest way to drive your cost up. All of our requests are cache aware, so we’re reusing a warm cache wherever possible. For example, our cache hit rate went from 5% → 60% in LibreChat once properly implemented.
Keep Context Lean – Start fresh sessions when switching tasks. Scope file context narrowly. Disconnect unused tools. Don't just compact. The goal isn't fewer tokens used, it's fewer tokens wasted.
Better Visibility – Our engineers can use as many tokens as they want, from whatever model they want, but we’ve made usage visible – and the more you spend on AI, the more impact we expect.
The goal isn't to suppress usage. It's to build the infrastructure that makes exponential growth sustainable.
Putting this into practice has cut our AI spend nearly in half, while our token usage continues to grow.
@feedly Hi! Since yesterday, I haven't been able to access my Feedly account—neither through the mobile app nor the web app. It treats me like a new user. Please let me know what information you need to verify my account. Thank you very much.
@feedly_support Hi! Since yesterday, I haven't been able to access my Feedly account—neither through the mobile app nor the web app. It treats me like a new user. Please let me know what information you need to verify my account. Thank you very much.
Este repositorio es una joya. Te da todos los pasos e instrucciones para proteger y asegurar tu servidor Linux.
Perfecto por si tienes un servidor propio o VPS:
https://t.co/yJD2GnXFPj
Potser us agradarà saber que s'està intentant engegar de nou Meteoprades gràcies a l'ajut de les Institucions.
No podem assegurar terminis, però hem cregut oportú comunicar-vos que hi ha una possible solució en marxa :)
🤞
37 milions i mig d'euros a projectes a Guatemala, Palestina o Moçambic, mentre a Catalunya un de cada 4 infants pateix risc de pobresa i centenars de malalts esperen tractaments vitals que el departament de sanitat considera "massa cars".
VERGONYA
Ajuda'm a arribar arreu!
Ok, to make up for the long time since the 2021 update, here's another change in a little less than an hour to fix some typos and add some notes.
The previous post already got a good number of reposts so again, version number to the rescue
Siempre he dicho que una sola ley cambiaría todo en España. Que el empresario ingrese el sueldo íntegro al trabajador… Y que el estado le quite el dinero al trabajador.
Nos vamos a reír y bien.
Màxima precaució en les pròximes 24 hores! Avís @meteocat màxim 6 sobre 6 intensitat torrencial al Baix Ebre i Montsià. A més del que ha caigut es podrien sumar més de 200 mm afegits. Situació de risc per les persones! Feu cas a equips emergències! @3CatInfoelTemps
Installation of Content
We're aware of an issue preventing players from accessing the game with some receiving an error stating they need to purchase DLC or similar.
Stay tuned for further updates as the team investigates this issue.
Thank you for your patience!