# Ada uses Cohere Embed on Model Vault for faster, cheaper retrieval

> Original title: Some great results in here!:

- Source: Aidan Gomez
- Published: 2026-10-27T17:17:45.000Z
- Bad Influence: https://badinfluence.news/items/o6f3my3gqa3cg3qriqudbin5f
- Original: https://x.com/aidangomez/status/2115130845069496380

## Summary

Ada achieved 77 ms median retrieval latency using Cohere Embed on Model Vault, according to Cohere's post. The setup delivered 2.7x greater throughput-per-dollar for offline embedding and a 4x smaller vector storage footprint with no measurable retrieval degradation. Ada also reached 90.7% top-10 recall.

## Post

Some great results in here!:

By using Cohere Embed on Model Vault, @ada_cx achieved:
* 77 ms median retrieval latency for faster, more accurate AI voice interactions
* 2.7x greater throughput-per-dollar for offline embedding, increasing cost efficiency
* 4x smaller vector storage footprint with no measurable retrieval degradation
* 90.7% top-10 recall for accurate information retrieval

https://cohere.com/customer-stories/ada