datasette-llm-usage 0.2a0

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

Summary

The `datasette-llm-usage` plugin has been updated to version 0.2a0, focusing on tracking LLM token usage within a SQLite database. This release removes features related to allowances and estimated pricing, delegating these functions to the new `datasette-llm-accountant` plugin. It now relies on `datasette-llm` for model configuration. A significant new feature allows logging full prompts, responses, and tool calls to the `llm_usage_prompt_log` table in the internal database, enabled by the `datasette-llm-usage.log_prompts` plugin setting. Additionally, the `/ -/llm-usage-simple-prompt` page has been redesigned and now requires specific `llm-usage-simple-prompt` permission.

Key takeaway

For developers managing LLM deployments with Datasette, you should upgrade to `datasette-llm-usage 0.2a0` to streamline token tracking. Be aware that allowance and pricing features have moved to `datasette-llm-accountant`, requiring a separate installation for those capabilities. Consider enabling prompt logging for enhanced debugging and auditing of LLM interactions.

Key insights

The `datasette-llm-usage` plugin now focuses solely on LLM token usage tracking, offloading accounting features.

Principles

• Modularize LLM accounting functions
• Centralize LLM model configuration

Method

Configure `datasette-llm-usage.log_prompts` to enable detailed logging of LLM interactions (prompts, responses, tool calls) to `llm_usage_prompt_log` table.

In practice

• Install `datasette-llm-accountant` for pricing
• Set `log_prompts` to true for full logging
• Grant `llm-usage-simple-prompt` permission

Topics

• datasette-llm-usage
• LLM Token Tracking
• SQLite Database
• Prompt Logging
• datasette-llm

Best for: NLP Engineer, AI Engineer, Machine Learning Engineer, MLOps Engineer

Related on AIssential

• datasette-llm-accountant 0.1a4 — datasette-llm-accountant 0.1a4 fixes a bug in tracking LLM response chains for Datasette.
• datasette-llm-limits 0.1a0 — The datasette-llm-limits plugin enables granular, periodic spending limits for LLM usage within Datasette.
• Token Economics: Why LLM Output Tokens Cost More Than Input Tokens — LLM output tokens cost more due to sequential generation and increasing memory bandwidth demands.
• On-Prem vs. Proxy Solutions for Secure LLM Usage: A Practical Guide for Enterprises — The deployment architecture for Large Language Models (LLMs) has become the defining factor for enterprise AI security, compliance, and sustainability, shifting focus from model capability to governance.
• LLM Cost Tracking Solution: How to Monitor and Control AI Spend in Agentic Systems — LLM cost tracking requires granular observability into token usage across complex agentic system execution paths.
• How to save up to 70% of tokens with LLMs? — Representing database schemas as TypeScript interfaces in LLM prompts significantly reduces token usage and improves accuracy.

Open in AIssential →

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