datasette-enrichments-llm 0.2a0

2026-04-01 · Source: Simon Willison's Weblog · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Data Science & Analytics, Software Development & Engineering · Depth: Intermediate, quick

Summary

The datasette-enrichments-llm plugin, version 0.2a0, has been released, introducing a significant change in how it manages and configures Large Language Models (LLMs). This update integrates with the datasette-llm plugin, allowing users to specify which LLM models are available for data enrichment tasks. The new configuration leverages the "enrichments" purpose within datasette-llm, providing more granular control over model selection. This release, posted on April 1st, 2026, streamlines the process of using LLMs to enhance datasets within the Datasette ecosystem, enabling more tailored and efficient data processing workflows.

Key takeaway

For Data Engineers and Analysts using Datasette for data processing, this update simplifies LLM integration. You can now precisely control which LLM models are accessible for enrichment tasks by configuring the "enrichments" purpose in datasette-llm, streamlining your data preparation workflows and ensuring appropriate model usage.

Key insights

datasette-enrichments-llm 0.2a0 integrates datasette-llm for enhanced LLM configuration and management.

Principles

• Modular plugin architecture
• Purpose-driven model configuration

Method

Configure LLM models for data enrichment by specifying the "enrichments" purpose within the datasette-llm plugin's settings.

In practice

• Specify available LLM models
• Tailor LLM use for data enrichment

Topics

• datasette-enrichments-llm
• datasette-llm
• Large Language Models
• Data Enrichment
• Plugin Configuration

Code references

• datasette/datasette-enrichments-llm
• datasette/datasette-llm

Best for: Machine Learning Engineer, NLP Engineer, Data Scientist, AI Engineer, Software Engineer

Related on AIssential

• datasette-enrichments-llm 0.2a1 — Passing the triggering actor to LLM enrichment methods enhances contextual control and personalization.
• datasette-llm 0.1a7 — Datasette-llm 0.1a7 enables configurable default options for LLMs used by dependent plugins.
• Models Can Model, But Can't Bind: Structured Grounding in Text-to-Optimization — LLMs' text-to-optimization performance is bottlenecked by accurately binding numerical data, not just modeling structure.
• How I use LLMs — LLMs are versatile tools, but understanding their underlying mechanisms and available features is key to effective use.
• Important LLM Papers for the Week From 22/12 To 28/12 — LLMs require specialized data preparation frameworks and robust scientific intelligence benchmarks to advance beyond isolated capabilities.
• DataShield: Safety-degrading Data Filtering for LLM Benign Instruction Fine-Tuning — DataShield efficiently identifies safety-degrading samples in benign LLM fine-tuning datasets by quantifying compliance behavior.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Simon Willison's Weblog.