> ## Documentation Index
> Fetch the complete documentation index at: https://docs.inference.net/llms.txt
> Use this file to discover all available pages before exploring further.
# Troubleshooting Training Failures
> Common training failures, what they look like, and how to recover.
Training runs should generally succeed without intervention. If a run fails, it usually indicates an underlying issue worth investigating.
## Common failures
| Failure | What happened | What to do |
| ------------------------------- | ----------------------------------------------------------------------------------- | ---------------------------------------------------------- |
| **Model collapse / divergence** | Model trained but performs poorly. May have overfit despite good eval scores. | Check data quality and diversity. Try a different recipe. |
| **Insufficient data** | Model fails to generalize well. | Add more representative samples to the training dataset. |
| **Poor data quality** | Training data isn't diverse or heterogeneous enough. | Curate a more varied dataset. Ensure it covers edge cases. |
| **Rate limited** | Too many concurrent training runs. The platform rate-limits usage to prevent abuse. | Wait for your current run to finish, then retry. |
## What you see on failure
* An error message on the training run
* A description of the failure condition
* A retry button
## When to reach out
If training fails unexpectedly, contact support. Failures usually indicate something worth looking into — the team can help diagnose whether it's a data issue, a recipe mismatch, or a platform problem.
Meet with our team to discuss training failures or get help with your approach.
## Training succeeded?
Model trained and eval scores look good? Deploy it to a dedicated GPU and start serving traffic.
Ship your trained model to production.