Setting a custom price for a model in AgentsView

2026-06-09 · Source: Simon Willison's Weblog · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Data Science & Analytics · Depth: Intermediate, quick

Summary

On June 9th, 2026, a method was detailed for setting custom model prices within AgentsView, Wes McKinney's Python toolkit for analyzing coding agent transcripts and token usage. This became necessary when the newly released Claude Fable 5 model was not yet included in AgentsView's default pricing database. The author reverse-engineered the tool to integrate custom pricing for Fable 5. An accompanying screenshot illustrates the cost attribution for Claude Fable 5 usage across various local projects, showing "prod_datasette_agent" as the dominant cost center at \$74.06, representing 89.3% of the total \$82.99 spent. The analysis also highlighted significant cache efficiency, reporting \$516.62 saved compared to uncached operations.

Key takeaway

For AI Engineers managing LLM agent deployments and seeking accurate cost attribution, you should anticipate the need to manually update pricing databases for newly released models like Claude Fable 5. This ensures AgentsView provides precise token usage and cost analysis across your projects. Proactively developing a process for integrating custom model pricing will maintain visibility into your operational expenses and highlight cache efficiency savings.

Key insights

AgentsView's model pricing database can be extended with custom rates for new or unsupported LLMs, enabling accurate cost tracking.

Principles

• Accurate LLM cost tracking is essential.
• Custom pricing extends analysis tools.
• Reverse engineering can adapt tools.

Method

Reverse-engineer AgentsView to identify and implement a custom pricing recipe for models not in its default database.

In practice

• Track token usage for new LLMs.
• Analyze cost attribution by project.
• Evaluate cache efficiency savings.

Topics

• AgentsView
• LLM Cost Management
• Token Usage Analysis
• Claude Fable 5
• Custom Model Pricing
• Coding Agents

Best for: MLOps Engineer, AI Engineer, Director of AI/ML

Related on AIssential

• Artificial Analysis: Independent LLM Evals as a Service — with George Cameron and Micah-Hill Smith — Independent benchmarking is crucial for navigating the complex, rapidly evolving AI model landscape and ensuring objective performance assessment.
• 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.
• Claude Code Heading for a $100 Paywall Instead of $20! The New Pricing Reality for AI Coding Agents — Advanced AI coding agent services are shifting from flat-rate, low-cost subscriptions to higher-priced tiers due to increased compute demands.
• Claude Fable 5 is here! — Claude Fable 5 offers unparalleled performance for complex, long-horizon tasks at a premium cost.
• Five things we learned trimming LibreChat’s LLM bill — Strategic LLM cost optimization requires understanding provider pricing, intelligent model routing, and careful resource management.
• Usage-based pricing killing your vibe - here's how to roll your own local AI coding agents — Local LLMs offer a cost-effective alternative to commercial coding agents, leveraging specific models and optimized inference.

Open in AIssential →

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