Understand this story
Aidan Gomez shared the measured results of ada_cx using Cohere Embed on Model Vault. According to his post published on October 28, 2026, the median retrieval latency was 77 milliseconds, throughput per dollar for offline embeddings increased 2.7 times, vector storage footprint shrank 4 times with no noticeable decline in retrieval quality, and top-10 recall reached 90.7%. Currently, these results come only from this single source, Aidan Gomez, with no other independent reports or subsequent developments.
AI-generated from public reports
Latest development · Report
Measured Results of Cohere Embed on Model VaultDevelopments
1 confirmed developments
Aidan Gomez shares results of applying Cohere Embed on Model Vault
Source:Aidan Gomez
Reports and discussion timeline
Browse reports and discussion by time. Their evidence levels remain distinct.
Oct 28
Aidan Gomez
Measured Results of Cohere Embed on Model Vault
Aidan Gomez shares the results of ada_cx using Cohere Embed on Model Vault: median retrieval latency of 77 ms, 2.7x higher throughput per dollar for offline embedding, 4x smaller vector storage footprint with no significant drop in retrieval quality, and a top-10 recall rate of 90.7%.