ai-spend
Your AI bills, in your terminal. No proxies. No dashboards. No SDK changes.
ai-spend pulls usage data directly from Anthropic and OpenAI's official billing APIs and displays it in your terminal. Your API keys stay on your machine. Your prompts are never seen.
$ ai-spend summary
Provider This Month vs Last Month
─────────────────────────────────────────
Anthropic $47.23 +12%
OpenAI $31.88 -4%
─────────────────────────────────────────
Total $79.11
Top models:
claude-opus-4-6 $31.44 (67% of Anthropic)
gpt-4.1 $18.22 (57% of OpenAI)
$ ai-spend daily --last 7
Date Anthropic OpenAI Total
──────────────────────────────────────────
2026-02-28 $8.12 $4.33 $12.45
2026-02-27 $6.88 $5.11 $11.99
...
How It's Different
Unlike gateway-based trackers, ai-spend never sees your prompts. It reads token counts and costs from your provider's billing API — the same data that appears on your invoice. Nothing more.
| Reads billing API directly | No proxy/SDK | Prompts stay local | CLI-native | |
|---|---|---|---|---|
| ai-spend | Yes | Yes | Yes | Yes |
| Langfuse | No (intercepts calls) | No | No | No |
| Helicone | No (proxy) | No | No | No |
| OpenAI dashboard | Yes | Yes | Yes | No (web only) |
Install
pip install ai-spend
Quick Start
# Add your providers (keys stored locally with 0600 perms)
ai-spend config add anthropic
ai-spend config add openai
# Pull latest usage data
ai-spend sync
# See your spend
ai-spend summary
ai-spend daily --last 30
Usage
# Provider management
ai-spend config add <provider> # Add a provider (anthropic, openai)
ai-spend config remove <provider> # Remove a provider
ai-spend config list # List configured providers
# Sync usage data from provider APIs
ai-spend sync
# View spend
ai-spend summary # Aggregated totals
ai-spend summary --json # JSON output
ai-spend daily # Daily breakdown
ai-spend daily --last 7 # Last 7 days
# Budgets
ai-spend budget set 100 # Set monthly budget ($100)
ai-spend budget check # Check against budget
# Manual entries (for providers without billing APIs)
ai-spend manual add 12.50 --provider ollama --note "local GPU costs"
# Export (Pro)
ai-spend export --format csv
ai-spend export --format json
# Status
ai-spend status # License tier + system info
How It Works
Provider APIs ──→ ai-spend sync ──→ Local SQLite
(Anthropic) (~/.ai-spend/spend.db)
(OpenAI) │
▼
ai-spend summary/daily/budget
- Configure — Add provider API keys (stored locally in
~/.ai-spend/config.yamlwith0600permissions) - Sync — Pulls usage records from official billing APIs into a local SQLite database
- Query — All read commands (
summary,daily,budget,export) hit the local database only
No background processes. No network calls except during sync. Run it in a cron job, pipe it into monitoring, or check it before standups.
Free vs Pro
| Feature | Free | Pro |
|---|---|---|
| Sync + summary + daily | Yes | Yes |
| Budget tracking | Yes | Yes |
| Manual entries | Yes | Yes |
| JSON output | Yes | Yes |
| CSV/JSON export | -- | Yes |
License
MIT
Metadata
Release files for ai-spend 0.2.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| ai_spend-0.2.0.tar.gz | 36.8 kB | With provenance | Details for ai_spend-0.2.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| ai_spend-0.2.0-py3-none-any.whl | Python 3 | none | any | Details for ai_spend-0.2.0-py3-none-any.whl |
Total release size: 62.2 kB
Release files / ai_spend-0.2.0.tar.gz
| Download URL | ai_spend-0.2.0.tar.gz |
|---|---|
| Size | 36.8 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
26055292ee485ffea231291656bcb4fbc095e601a0e3a45ad00b735ef4280f82
|
|
BLAKE2b-256 checksum How to use checksums |
aefb3c4f3eff92440ef87f9ccb403664636de1c7cd90edb6f9cc9d37525be611
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
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 Mar 8, 2026.
Transparency logRelease files / ai_spend-0.2.0-py3-none-any.whl
| Download URL | ai_spend-0.2.0-py3-none-any.whl |
|---|---|
| Size | 25.4 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
0ab1c3636acab72eda2576162a9a51d28d1667367e6f0e637de7ba51542e977b
|
|
BLAKE2b-256 checksum How to use checksums |
a5eae4b26fd7e0b7b07da91092233e9c0988a7f711144aada934eae724168678
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
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 Mar 8, 2026.
Transparency logSecurity for ai-spend 0.2.0
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.
Full provenance
All 2 files in this release have been uploaded with provenance metadata.
Attestation for ai_spend-0.2.0.tar.gz
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 Mar 8, 2026.
Transparency logA note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.