### Check for existing issues

- [x] I have searched the existing issues and checked that my issue is not a duplicate.

### What happened?

When exporting usage for the same date range, daily_with_models shows much higher totals than daily and daily_with_keys. Example from our data:

daily / team total tokens: ~1.339B daily_with_models total tokens: ~34.275B

### Steps to Reproduce

1. Open Usage export in LiteLLM dashboard. 2. Select the same date range and same filters. 3. Export as daily and note total tokens/spend. 4. Export as daily_with_keys and compare totals (should align with daily). 5. Export as daily_with_models and sum tokens/spend. 6. Observe daily_with_models is significantly larger than daily/daily_with_keys.

### Relevant log output

```shell No backend error logs. Observed output mismatch from exports:

daily total_tokens: 1,339,000,000 (approx) daily_with_models total_tokens: 34,275,000,000 (approx) ```

### What part of LiteLLM is this about?

UI Dashboard

### What LiteLLM version are you on ?

v1.83.7-stable

### Twitter / LinkedIn details

_No response_