![](https://crypto4nerd.com/wp-content/uploads/2024/04/0Zioo3uB-jJJL7Ktd.png)
We spend a lot of time speaking to AI engineers every week. Some of the pain points we hear about very frequently include:
- “I don’t have a clear view of model performance over time”
- “I don’t know why the model is not performing well”
- “I can’t identify or prevent model quality regressions”
- “I don’t know which prompt / model is better
We built a new Analytics dashboard to give users new ways to slice and dice metrics. Here’s what you can do with it 👇
- Comparison reports of evaluation metrics and usage metrics
Compare latency, cost, feedback, and evaluation metrics across different prompts, models, customer ids, topics, and date ranges. Here’s a product video:
https://athina.ai/videos/compare-view.mp4
2. Analytics Dashboard
- Percentile distributions graph for every evaluation metric
- Graph of any evaluation metric over time
- New views and filters for token usage, cost, latency, user feedback
(click the image to see a 30 second product demo)
https://athina.ai/videos/eval-analytics.mp4
Without measurement, it’s very difficult to improve model performance.
Athina’s new LLM Analytics platform can help you:
- Identify regressions in models (across any criteria)
- Deep dive into the metrics segmented by prompt, customer, topic or model to discover gaps in model performance.
- Explore the performance of models over time
- Existing users: Sign in to Athina and access the Analytics and Compare dashboards.
- New users: Sign up, and log inferences in 5 mins. Then access the Analytics and Compare dashboards.
As always if you have any questions, feel free to reply to this email or get on a call with us!