Does q_auto create multiple files like f_auto does?

chaitanya_dasa Member Posts: 2
edited December 2023 in Digital Asset Management

I want to clarify my transformation expenditure when using q_auto. As far as I understand, f_auto creates four different formats of the asset to be served depending on the compatibility of the end-user browser client. I also understand that q_auto analyses each individual asset to determine the optimal delivery quality.

Consider a scenario where client_1 requests an asset and it is determined that the delivery quality should be low. The transformation is processed and client_1 receives a low quality file. Is this file then cached so if client_2 comes along and also requires a low quality asset, the cached file will be served instead of processing a new transformation?


Best Answer

  • DannyFromCloudinary
    DannyFromCloudinary Member, Cloudinary Staff Posts: 105
    Answer ✓

    Hey @chaitanya_dasa.

    When you request an asset with q_auto, rather than making separate renders for low/medium/high quality for each individual format, we generate a single q_auto render which aims to have the excellent visual fidelity, while still applying some compression, metadata stripping, etc.

    So, if you were to use f_auto,q_auto, this would essentially result in creating renders for f_jpg,q_auto, f_png,q_auto, f_avif,q_auto, etc, dynamically as needed

    For instance, if user A comes requests an asset via Google Chrome on desktop and we determine that a webp is the best format to use, we will deliver f_webp,q_auto, costing one transformation. If another Chrome user requests that same image, we will deliver it again, this time with no transformation cost as the image has already been generated. If user C comes along with a different browser and we determine that png is the best format to use, we'll create an f_png,q_auto render and deliver that

    I hope this helps, but if anything is unclear, or you have any further questions, please don't hesitate to ask.

    Kind regards,



  • chaitanya_dasa
    chaitanya_dasa Member Posts: 2

    Hey @DannyFromCloudinary ,

    Thank you for the reply.

    So, when paired with f_auto (as I have it in my configuration), q_auto is only processed once for every file format and then subsequent requests are served from CDN cache.

    Awesome. Thanks for the clarification.

  • Cloudinary Team
    Cloudinary Team Administrator, Cloudinary Staff Posts: 129 admin

    Hi there,

    That is correct 😊

    Let us know if you have any additional questions.

    Kind regards,


    Helpful Links For You
    💬 Share questions, connect with other users in our Cloudinary Community forums and Discord server!
    🧑‍🎓 Join our Cloudinary Academy for free courses, workshops and other educational resources.
    📄 Read our documentation for in-depth details on Cloudinary product features and capabilities
    📰 Check out the Cloudinary blog for the latest company news and insights