Databricks rolls out Astra model to engineers
源:https://t.co/ykphYTsAmW
For developers, Astra is a new tool that can boost productivity on complex coding tasks, though it comes with a higher cost than simpler models.
Databricks has rolled out its new Astra model to 3,500 engineers. Astra outperforms previous models on complex tasks like system design. Engineers using Astra increased coding spend by 60% compared to a baseline. The company used a pilot with 200 users to test the model before the full rollout.
源:https://t.co/ykphYTsAmW
源: x.com/pwendell/statu… Patrick Wendell @pwendell Today we rolled out Astra to every engineer at Databricks (N=~3500). Some notes that may be helpful to others: 1. Astra unambiguously out performs our previous highest-end models (Opus 5, Sol 5.6) on highly complex tasks, especially those related to high level system design or long range horizontal tasks. 2. Engineers given Astra increased overall coding spend by around 60% compared to baseline. 3. It is not clear Astra meaningfully improves on medium/low complexity coding tasks compared to earlier models. We suspect those tasks are mostly saturated (i.e. perfectly executed) by existing models. 4. We learned above by piloting Astra with around 200 users to gain signal on both quality and cost. We use Unity Gateway to do cohort-based experiments for all new models. 5. We give engineers a sub-budget specific to Astra to encourage them to use Astra selectively on complex tasks while preferring lower cost models for everyday tasks. Our engineers are able to mix-and-match tools and models within their overall budget envelope (we also allow for increased budgets through various mechanisms). These budgets are defined in Unity Gateway and regularly revisited. Note: We do not have robust comparisons of Astra-vs-Fable because we have net yet rolled out Fable widely due to data retention policies. 🔗 View Quoted Tweet 💬 0 🔄 0 ❤️ 0 👀 221 ⚡
