<?xml version="1.0" encoding="UTF-8"?><oembed><type>video</type><version>1.0</version><html>&lt;iframe src=&quot;https://www.loom.com/embed/c58e47dccf1441f9970d48b41cc32b3e&quot; frameborder=&quot;0&quot; width=&quot;1920&quot; height=&quot;1440&quot; webkitallowfullscreen mozallowfullscreen allowfullscreen&gt;&lt;/iframe&gt;</html><height>1440</height><width>1920</width><provider_name>Loom</provider_name><provider_url>https://www.loom.com</provider_url><thumbnail_height>1440</thumbnail_height><thumbnail_width>1920</thumbnail_width><thumbnail_url>https://cdn.loom.com/os/loom-media-content/loom/f20734ba-53dd-4b68-8c8b-942cb6a9dfb4/827b500d-5243-4144-8672-b02651eb01f8/sessions-thumbnails-c58e47dccf1441f9970d48b41cc32b3e-3d238e4ce3d782eb.gif/binary?authToken=eyJraWQiOiJtaWNyb3MtZ3JvdXAvdGRwLW9zLWFyY2hldHlwZS9wOXEwa3BuN3JpaGluZGk1IiwiYWxnIjoiUlMyNTYifQ.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.kNtEWZ3F1COIIUYjoSEVMAXopyBt17UU-I8j9bf8YXGy0UqAI3Yf1ALhiHQnGb3VrdyG5Z1c9XIm_bDdrnb_tYvawsch7e4tkaHzqU2WA55ZDobAT1vSRdpOCtRqZ1z-LMLjgX_t0vVboWVacEi4I1Gr6FRMmuhcSuGI-njCf_HvVPHQL5zVXuwXlApS2M3Y2LguWWAT9XiOwAHxR1QGgMr5Cz6cse9w_3lQLGkY2HbFTPJV6bX5h5AjqVbUu0TI46t_F_CAw_-I8GGPKw3gctusQd9qp33bvTBQm6-ZENyRx_D5HpL52heX7IEcz7rdBTVSTwNDD3v-FH-dAaujmg&amp;Expires=1795081973&amp;Signature=iPACL3Pq8TPIbwYT2iWH8lVED8JltZ-XQ6y7KLXYeHJ1q-BY98HvapU59RxPc7g2rW4q7tdkQ1ODDk0%7E7HCjuNnk-URT1KAaxBKs1es9L2A%7EKfo-0GkyWG6ZWuN2qUldMiHm2HICvNRFLoiewuw%7EOmgObuK3rGj6MZhHIh0%7E6i1py8tfNXypo5%7Eja1W1muIQo48nibHvHnWpQLWpn8FGu2LBbHavzIh5aA-XR1hUcuvHuGgLskf2TLhhSSK0Q9PdDjtvDiwTituRWwToIiec8sO%7Eic3ioWACPcyr6M0RnKIbst3seTONvHTgYYjhKRynywuJnLwl47iVBi2su0%7E1sg__&amp;Key-Pair-Id=KQOSYIR44AIC0</thumbnail_url><duration>60.533333</duration><title>Fine Tune Open Models Without Glue Code</title><description>This Loom discusses why a fine-tuning platform like Tensurant helps teams avoid engineering overhead. It says instead of writing training scripts, debugging Kuda errors, and building evaluation harnesses and glue code, you can fine-tune while monitoring loss and evaluating every run against a base model to see measurable improvements. The video also notes that after fine-tuning, you can publish and test in the playground, then serve for inference through an OpenAI compatible API. It further claims training runs on the GPU provider you already use, keeps data and weights in your own account, and aims to reduce GPU costs.</description></oembed>