# Training new model fails every time

**URL:** <https://discuss.roboflow.com/t/training-new-model-fails-every-time/9300>\
**Category:** 🤝  Community Help\
**Created:** [April 2, 2025, 6:12pm UTC](https://discuss.roboflow.com/t/training-new-model-fails-every-time/9300 "2025-04-02T18:12:48Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![Wolphy](https://yyz1.discourse-cdn.com/flex029/user_avatar/discuss.roboflow.com/wolphy/32/7040_2.png) [@Wolphy](https://discuss.roboflow.com/u/Wolphy)\
**Post date:** [April 2, 2025, 6:12pm UTC](https://discuss.roboflow.com/t/training-new-model-fails-every-time/9300/1 "2025-04-02T18:12:48Z")

</div>

We are trying to train new versions of our semantic segmentation model but have had them fail repeatedly. We tried creating an identical version (same settings) as one we created a few weeks ago that worked fine, but have the same issue with training failing after a few minutes. At this point we believe this is likely some other error, any help would be very much appreciated.

For reference we initially thought it was to do with image sizes, so attempted a few versions with reduced sizes, but none of these solved the problem and the earlier successful version did not have any of these problems

- **Project Type:** Semantic Segmentation
- **Operating System & Browser:** windows 10 - Chrome
- **Project Universe Link or Workspace/Project ID:** path-segmentation-iwp4c

---

<div class="post-metadata">

**Author:** ![system](https://canada1.discourse-cdn.com/flex029/uploads/roboflow1/original/1X/bb6edbfa56d0cf2ee603f061e9e09103502cc8cc.png) [@system](https://discuss.roboflow.com/u/system)\
**Post date:** [April 23, 2025, 6:13pm UTC](https://discuss.roboflow.com/t/training-new-model-fails-every-time/9300/2 "2025-04-23T18:13:22Z")

</div>

Closed on April 23, 2025, 6:13pm UTC

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.