burn-o-meter
See what your AI coding agents really cost. Tokens, spend, cache efficiency and rate limits for Claude Code, Codex, OpenCode and Kimi Code — read from the logs they already write on your machine. No account, no telemetry, nothing sent anywhere.
Install
pipx install burn-o-meter # or: uv tool install burn-o-meter
Then:
burn-o-meter scan && burn-o-meter today
burn-o-meter statusline install # Claude's rate-limit %, live from Claude Code
Requires Python 3.11+. The macOS menu bar app is a separate, optional build — see installing.
What it gets right
Every figure is checked against real logs, because each agent's format hides a trap that documentation never mentions:
- Claude Code writes the same message up to 7 times — summing naively overcounts by about 2.5×. Subagents keep their turns in separate files, and write a message while it streams.
- Codex repeats its final usage event, and restates each turn in a second record.
- A published rate of
0means "included in your plan", not "free"; a model on your own hardware is not metered, never$0.00. - Ollama truncates prompts past its context window and reports what it read.
Costs on a subscription are labelled API-equivalent, never presented as a bill, and a model with no known rate shows as unpriced rather than guessed.
Private by construction
It never reads prompts or completions, never opens credential files, and makes no
network call except burn-o-meter pricing refresh, which you run yourself. Each
of those is enforced by tests rather than promised —
SECURITY.md
lists them.
1 maintainer
Metadata
1 maintainer
Release files for burn-o-meter 0.6.5
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| burn_o_meter-0.6.5.tar.gz | 2.5 MB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| burn_o_meter-0.6.5-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 2.6 MB
Release files / burn_o_meter-0.6.5.tar.gz
| Download URL | burn_o_meter-0.6.5.tar.gz |
|---|---|
| Size | 2.5 MB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
b94f049b88dcf5fea6fa51efa0718992a6462da4d726b195a771fe3c6aa49f9a
|
|
BLAKE2b-256 checksum How to use checksums |
42ce24a8cb557167d8e61aa78b78262424fa3ec4d9653e0ae80733b5ea8300f7
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 28, 2026.
Transparency logRelease files / burn_o_meter-0.6.5-py3-none-any.whl
| Download URL | burn_o_meter-0.6.5-py3-none-any.whl |
|---|---|
| Size | 112.0 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
59951dca455c6c75b3f7e03d6eacfcfec4772a8ab50c1f2aecb8c48e454185b5
|
|
BLAKE2b-256 checksum How to use checksums |
c80d0f1129da2d6def684423b711d5548cc3696f2c203479d8e6761ca5792482
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 28, 2026.
Transparency log
