Never get throttled mid-task again

Open-source background agent that tracks your Synthetic and Z.ai API quota usage across every reset cycle. Historical trends, rate projections, per-session tracking, and a Material Design 3 dashboard. Single binary, ~25 MB RAM, zero telemetry. Works with Cline, Roo Code, Kilo Code, Claude Code, and any tool using Synthetic or Z.ai API keys.

One-command install
$ curl -fsSL https://raw.githubusercontent.com/onllm-dev/syntrack/main/install.sh | bash Copy

v1.6.0 · Go · GPL-3.0 License · Zero Telemetry

localhost:9211
Dashboard
Synthetic Z.ai Both
Subscription
3h 47m
0%
154 / 1,350
Search (Hourly)
38m
238 / 250
Tool Calls
16h 22m
0%
14,191 / 16,200
Usage over time — Last 6 hours
Subscription Search Tool Calls

Supported Providers

S
Synthetic
synthetic.new
Z.ai
z.ai
Both View
Side-by-side comparison
Anthropic Coming Soon
anthropic.com

Telemetry your API provider doesn't give you

SynTrack fills the gap between "current usage snapshot" and the historical, per-cycle, cross-session intelligence developers actually need.

Historical Usage Trends

Chart.js time-series with 1h, 6h, 24h, 7d, and 30d ranges. See patterns your provider's dashboard never shows.

Reset Cycle Detection

Automatically detects when quotas reset by tracking renewsAt changes. Per-cycle peak, total delta, and average rate.

Live Countdown Timers

Real-time countdown to each quota reset. Color-coded status badges — healthy, warning, danger, critical.

Rate Projections

Current consumption rate extrapolated to reset time. Know if you'll hit the limit before it resets.

Session Tracking

Every agent run logs peak consumption. Compare usage across coding sessions — "this session used 342 requests."

Usage Insights

Provider-specific stats: cycle utilization, billing period usage, weekly pace, tool call share, tokens-per-call efficiency.

How it works

A background agent polls your API quotas, stores snapshots in SQLite, and serves a dashboard. That's it.

API Providers
Synthetic + Z.ai
SynTrack Agent
Background polling
Dashboard
:9211
~25 MB
RAM Idle
0
Dependencies
1
Binary
60s
Poll Interval

Get up and running in minutes

Choose the method that works best for you. Supports macOS (ARM64, AMD64) and Linux (AMD64, ARM64).

Manual Download

Download the binary for your platform from GitHub Releases, configure your .env, and run directly.

bash
# Download (Linux AMD64 example) curl -L -o syntrack \ https://github.com/onllm-dev/syntrack/releases/latest/download/syntrack-linux-amd64 chmod +x syntrack # Create config cat > .env << 'EOF' SYNTHETIC_API_KEY=syn_your_key_here ZAI_API_KEY=your_zai_key_here SYNTRACK_ADMIN_USER=admin SYNTRACK_ADMIN_PASS=changeme SYNTRACK_PORT=9211 EOF # Run ./syntrack # start (background) ./syntrack --debug # foreground mode ./syntrack stop # stop ./syntrack status # check status

Want to build from source? See DEVELOPMENT.md

What your provider doesn't show you

API providers show current usage. SynTrack shows everything else.

Capability Synthetic Z.ai SynTrack
Current quota usage
Reset time visibility
Historical usage trends
Reset cycle detection & history
Per-cycle consumption stats
Usage rate & projections
Per-session tracking
Multi-provider unified view
Live countdown timers
Open source & self-hosted

Who is SynTrack for?

Whether you're a solo developer tracking your $20/month Synthetic budget or a team lead managing shared API keys, SynTrack gives you the data your provider doesn't.

Solo Developers & Freelancers

Using Cline, Roo Code, or Kilo Code with a Synthetic or Z.ai API key? SynTrack tracks your quota burn rate so you never get throttled mid-task. Know exactly where your AI coding budget goes.

Small Dev Teams (3-20 people)

Sharing API keys across a team? SynTrack provides a shared dashboard with historical trends, session tracking, and rate projections. Budget with data, not guesses. Present usage reports in sprint reviews.

DevOps & Platform Engineers

Deploy SynTrack as a lightweight sidecar (~25 MB) with no SaaS dependency. Use the SQLite database as a Grafana data source or the REST API endpoints to feed your existing monitoring stack.

Privacy-Conscious Developers

You chose Synthetic because it never trains on your data. SynTrack has the same philosophy: single binary, local SQLite, zero telemetry. Works in air-gapped and regulated environments.

Researchers & Educators

Track per-session API consumption for grant reports and paper methodology sections. Compare Synthetic vs Z.ai efficiency. Z.ai's $3/month plan is popular with students and researchers.

Budget-Conscious API Users

At $20-$60/month on Synthetic or $3-$15/month on Z.ai, every request matters. SynTrack shows if you're underutilizing your plan, burning through quota too fast, or need to upgrade.

Frequently asked questions

How do I track my Synthetic API usage?

Install SynTrack with one command: curl -fsSL https://raw.githubusercontent.com/onllm-dev/syntrack/main/install.sh | bash. Set your SYNTHETIC_API_KEY in the .env file and run ./syntrack. SynTrack polls the Synthetic /v2/quotas endpoint every 60 seconds (configurable), stores historical data in SQLite, and serves a dashboard at localhost:9211 showing subscription, search, and tool call quotas with live countdowns and rate projections.

How do I monitor Z.ai (GLM Coding Plan) API quota?

Set your ZAI_API_KEY in the .env file. SynTrack polls the Z.ai /monitor/usage/quota/limit endpoint and tracks token limits, time limits, and tool call quotas. Both Synthetic and Z.ai can run simultaneously with a unified "Both" view on the dashboard.

Does SynTrack work with Cline, Roo Code, Kilo Code, or Claude Code?

Yes. SynTrack monitors the API provider (Synthetic or Z.ai), not the coding tool. Any tool that uses a Synthetic or Z.ai API key—including Cline, Roo Code, Kilo Code, Claude Code, Cursor, Windsurf, and others—will have its usage tracked automatically.

Is SynTrack a cross-platform alternative to Quotio?

Yes. Quotio is macOS-only (requires macOS 15.0+) and tracks Claude, Gemini, and OpenAI. SynTrack runs on macOS, Linux, and Windows, and tracks Synthetic and Z.ai specifically. SynTrack also provides historical usage trends, reset cycle detection, per-session tracking, and a web-based dashboard that Quotio does not offer.

How is SynTrack different from Helicone or Langfuse?

Helicone ($25/mo+) and Langfuse (Docker/SaaS) are LLM observability platforms designed for custom AI pipelines with per-token cost tracking. SynTrack is a single 25 MB binary focused on tracking flat-rate AI coding API subscriptions. No SaaS accounts, no Docker, no proxy setup—just download and run.

Does SynTrack send any data to external servers?

No. SynTrack has zero telemetry. All usage data is stored locally in a SQLite file on your machine. The only outbound network calls are to the Synthetic and Z.ai quota APIs you configure. No analytics, no tracking, no cloud. The source code is fully auditable on GitHub (GPL-3.0).

How much memory does SynTrack use?

SynTrack idles at ~25-30 MB RAM and peaks at ~50 MB during dashboard rendering. Breakdown: Go runtime (5 MB), SQLite in-process (2 MB), HTTP server (1 MB), polling buffer (1 MB). This is lighter than a single browser tab and designed to run as a background daemon indefinitely.

What platforms does SynTrack support?

Pre-built binaries are available for macOS (ARM64 and AMD64), Linux (AMD64 and ARM64), and Windows (AMD64). SynTrack is written in pure Go with no CGO dependencies, so it cross-compiles cleanly to any Go-supported platform.

Start tracking your API usage

SynTrack is free, open source, and takes under a minute to set up. No accounts, no cloud, no telemetry sent anywhere. Your data stays on your machine.