Prices every LLM call from the logs you already have.
105 models built in. Prompt caching priced right.
- Reads OpenAI, Anthropic and Claude Code logs
- Tracks a budget and forecasts the month
- Flags calls that cost far more than usual
- Serves Prometheus metrics
- Runs in the terminal or as a Rust library
cargo install --locked llm-cost-dashboardllm-dash --demo
llm-dash --log-file requests.ndjson --budget 50More: docs/DETAILS.md
License: MIT