Datasette-LLM Adds Plugin Hook For Model Purposes
Datasette-LLM 0.1a1 was released as a base plugin that exposes LLM models for other Datasette plugins to use, including datasette-enrichments-llm. It adds register_llm_purposes() and get_purposes() hooks and lets plugin authors map named purposes (for example, 'enrichment' → GPT-5.4-nano, 'sql-assist' → Sonnet 4.6), call llm.model(purpose='enrichment'), and enable admin UIs to assign models to purposes.
Key Points
- 1Introduces register_llm_purposes() and get_purposes() hooks for registering and retrieving purpose strings.
- 2Enables centralized model-to-purpose mapping so different plugins use appropriate LLMs for specific tasks.
- 3Allows plugin developers to call llm.model(purpose='enrichment') and power admin UIs for assignments.
Scoring Rationale
Official, practical plugin enhancement with direct developer value; limited impact outside the Datasette plugin ecosystem.
Sources
Public references used for this report.
Practice with real Logistics & Shipping data
90 SQL & Python problems · 15 industry datasets
250 free problems · No credit card
See all Logistics & Shipping problems
