otelcol-genai-sketches
by Community
Track LLM token usage by user, session and model without a metric label for every user. Prometheus or any OTLP metrics backend.
OSS
otelcol-genai-sketches
Added 16 Oct 2026
Overview
A Go-based OpenTelemetry collector component that tracks LLM token usage by user, session, and model without creating a metric label for every user. It exports metrics to Prometheus or any OTLP metrics backend.
Best for
Best for
Teams needing low-cardinality LLM token usage tracking in existing Prometheus or OTLP stacks
Use cases
- Monitor token consumption per user and session
- Export LLM usage metrics to Prometheus
- Track model-level token spend in OTLP pipelines
Notes
5 stars on GitHub. Last updated 2026-10-16. Licensed Apache-2.0.
Pros
- Avoids high-cardinality labels for per-user metrics
- Works with standard Prometheus and OTLP backends
- Open source and written in Go
Cons
- Very early stage with only 5 GitHub stars
- Limited community and documentation
- Requires self-hosting and configuration
Get the free Developer’s Field Guide
A 27-page field guide to the AI coding workflow with Claude. Claude Code, MCP servers, the prompt patterns that work, and what to delegate. Free.
Enter your work email. We send it straight over, plus a few short notes worth knowing. Unsubscribe any time.