Databricks releases Enterprise AI Usage Index
Understand this story
Databricks has released a public resource, the Databricks AI Index, which aggregates anonymized token consumption data from more than 10,000 enterprises on its Unity Gateway and publishes coding-agent evaluations based on an internal private codebase, updated weekly. Databricks says this is the first time it has permanently published its internal evaluation suite, used to assess newly released closed-source and open-weight models. The current dataset covers September 21 to November 7, 2026. The index presents both actual enterprise usage and coding-agent evaluation information.
AI overview based on source reports · Updated 39 min ago
Coverage timeline
Follow the coverage to explore different perspectives.
- Databricks BlogIntroducing the Databricks AI Index: The state of enterprise AI tokenomics
Databricks launched the Databricks AI Index, a public resource combining aggregated, anonymized usage data from 10,000+ companies on Unity Gateway with weekly-updated coding-agent benchmarks drawn from real workflows in its private codebase. The Index reports that OpenAI's enterprise token share grew on its 6 series Sol and Astra models, while Claude Opus 5.5 rose from 7.2% to 37.9% of proprietary-model tokens in six weeks, and open-weight models grew from 7.3% in July to 30.9% in November. It also notes that cost per completed task can diverge sharply from token price—GPT-6 Luna and Claude Opus 5.5 both hit 89.1% task success at $0.05 versus $1.62 per task—and that median weekly user spending more than doubled from $1.10 while the 99th-percentile user spent $699.74 versus a $2.49 median.
Activity for this story
Not enough continuous observations to show a trend yet.