Fine-tuning with the Cohere dashboard

Customers can kick off fine-tuning jobs by completing the data preparation and validation steps through the Cohere dashboard. This is useful for customers who don't need or don't want to create a fine-tuning job programmatically via the Fine-tuning API or via the Cohere Python SDK, instead preferring the ease and simplicity of a web interface.


Before a fine-tune job can be started, users must upload a dataset with training and (optionally) evaluation data. The contents and structure of the dataset will vary depending on the type of fine-tuning. Read more about preparing the training data for Chat, Classify, and Rerank fine-tuning.

Starting a Fine-tune job

After uploading the dataset and going through the validation and review data phases in the UI, the fine-tune job can begin. Read more about starting the fine-tuning jobs for Chat, Classify, and Rerank.

Fine-tune results

You will receive an email notification when the fine-tune model is ready. You can explore the evaluation metrics using the Dashboard and try out your model using one of our APIs on the interactive Playground.

Fine-tune job statuses

As your fine-tune job progresses, it will progress through various stages. The following table describes the meaning of the various status messages you might encounter:

QueuedFine-tune job is queued to start.
CreatedFine-tune job has been created and will start training soon.
TrainingFine-tune job is currently training.
DeployingFine-tune job has finished training and is deploying the model endpoint.
ReadyFine-tune job has finished and is ready to be called.
AsleepFine-tune job has been paused. To wake the model, you can select the "Wake Model" button on the Status tab.
FailedFine-tune job has failed. Please contact customer support if you need more help in understanding why your fine-tune has failed.