OpenRouter Usage Analytics

Overview

OpenRouter provides usage data through three endpoints: GET /api/v1/auth/key (credit balance and rate limits), GET /api/v1/generation?id= (per-request cost and metadata), and response usage fields (token counts). This skill covers collecting metrics from these sources, building analytics pipelines, cost reporting, and performance dashboards.

Prerequisites

• An OpenRouter API key (sk-or-v1-...) exported as OPENROUTERAPIKEY — see the openrouter-install-auth skill for setup
• Python 3.8+ with the OpenAI SDK plus requests (used to fetch exact per-request cost from the generation endpoint); sqlite3 (stdlib) backs the analytics database
• curl and jq for the Credit Balance Monitoring one-liner
• HTTP-Referer / X-Title headers set on the client if you also want OpenRouter dashboard attribution

Instructions

1. Route completions through trackedcompletion (Collect Per-Request Metrics) — it times each call, then fetches the exact totalcost from GET /api/v1/generation?id= and emits a JSON metric with tokens, latency, and modelrequested vs modelused.
2. Initialize openrouteranalytics.db with initanalyticsdb (Analytics Database) and persist every metric via storemetric — the generation_id unique constraint plus INSERT OR IGNORE deduplicates retries.
3. Query the store with the Analytics Queries: daily cost summary, cost by model, top users by spend, hourly request pattern, and 30-day cost trend.
4. Watch remaining credits with the Credit Balance Monitoring snippet — curl + jq against /api/v1/auth/key reports creditsused, creditlimit, and remaining.
5. Generate the Weekly Report Generator output for stakeholders (totals plus top 5 models by cost).
6. Apply the retention and alerting policies from Enterprise Considerations (aggregate raw rows after 30 days, alert when daily cost exceeds 2x the historical average).

Collect Per-Request Metrics

import os, time, json, logging
from datetime import datetime, timezone
from openai import OpenAI
import requests as http_requests

log = logging.getLogger("openrouter.analytics")

client = OpenAI(
    base_url="https://openrouter.ai/api/v1",
    api_key=os.environ["OPENROUTER_API_KEY"],
    default_headers={"HTTP-Referer": "https://my-app.com", "X-Title": "my-app"},
)

def tracked_completion(messages, model="openai/gpt-4o-mini", user_id="system", **kwargs):
    """Make a completion and capture full analytics."""
    start = time.monotonic()
    response = client.chat.completions.create(
        model=model, messages=messages, **kwargs
    )
    latency = (time.monotonic() - start) * 1000

    # Fetch exact cost from generation endpoint
    cost = 0.0
    try:
        gen = http_requests.get(
            f"https://openrouter.ai/api/v1/generation?id={response.id}",
            headers={"Authorization": f"Bearer {os.environ['OPENROUTER_API_KEY']}"},
            timeout=5,
        ).json()
        cost = float(gen.get("data", {}).get("total_cost", 0))
    except Exception:
        pass

    metric = {
        "timestamp": datetime.now(timezone.utc).isoformat(),
        "generation_id": response.id,
        "model_requested": model,
        "model_used": response.model,
        "prompt_tokens": response.usage.prompt_tokens,
        "completion_tokens": response.usage.completion_tokens,
        "total_cost": cost,
        "latency_ms": round(latency, 1),
        "user_id": user_id,
    }
    log.info(json.dumps(metric))
    return response, metric

Analytics Database

import sqlite3

def init_analytics_db(db_path: str = "openrouter_analytics.db"):
    conn = sqlite3.connect(db_path)
    conn.execute("""
        CREATE TABLE IF NOT EXISTS metrics (
            id INTEGER PRIMARY KEY AUTOINCREMENT,
            timestamp TEXT NOT NULL,
            generation_id TEXT UNIQUE,
            model_requested TEXT,
            model_used TEXT,
            prompt_tokens INTEGER,
            completion_tokens INTEGER,
            total_cost REAL,
            latency_ms REAL,
            user_id TEXT
        )
    """)
    conn.execute("CREATE INDEX IF NOT EXISTS idx_metrics_ts ON metrics(timestamp)")
    conn.execute("CREATE INDEX IF NOT EXISTS idx_metrics_model ON metrics(model_used)")
    conn.execute("CREATE INDEX IF NOT EXISTS idx_metrics_user ON metrics(user_id)")
    conn.commit()
    return conn

def store_metric(conn, metric: dict):
    conn.execute(
        """INSERT OR IGNORE INTO metrics
           (timestamp, generation_id, model_requested, model_used,
            prompt_tokens, completion_tokens, total_cost, latency_ms, user_id)
           VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?)""",
        (metric["timestamp"], metric["generation_id"], metric["model_requested"],
         metric["model_used"], metric["prompt_tokens"], metric["completion_tokens"],
         metric["total_cost"], metric["latency_ms"], metric["user_id"]),
    )
    conn.commit()

Analytics Queries

-- Daily cost summary
SELECT date(timestamp) as day,
       COUNT(*) as requests,
       SUM(prompt_tokens + completion_tokens) as total_tokens,
       ROUND(SUM(total_cost), 4) as total_cost,
       ROUND(AVG(latency_ms)) as avg_latency_ms
FROM metrics
WHERE timestamp > datetime('now', '-7 days')
GROUP BY day ORDER BY day DESC;

-- Cost by model (this week)
SELECT model_used,
       COUNT(*) as requests,
       ROUND(SUM(total_cost), 4) as cost,
       ROUND(AVG(latency_ms)) as avg_ms,
       SUM(prompt_tokens) as total_prompt,
       SUM(completion_tokens) as total_completion
FROM metrics
WHERE timestamp > datetime('now', '-7 days')
GROUP BY model_used ORDER BY cost DESC;

-- Top users by spend
SELECT user_id, COUNT(*) as requests,
       ROUND(SUM(total_cost), 4) as total_cost,
       ROUND(AVG(total_cost), 6) as avg_cost_per_request
FROM metrics
WHERE timestamp > datetime('now', '-30 days')
GROUP BY user_id ORDER BY total_cost DESC LIMIT 20;

-- Hourly request pattern (for capacity planning)
SELECT strftime('%H', timestamp) as hour,
       COUNT(*) as requests,
       ROUND(AVG(latency_ms)) as avg_latency
FROM metrics
WHERE timestamp > datetime('now', '-7 days')
GROUP BY hour ORDER BY hour;

-- Cost trend (daily, last 30 days)
SELECT date(timestamp) as day, ROUND(SUM(total_cost), 4) as cost
FROM metrics
WHERE timestamp > datetime('now', '-30 days')
GROUP BY day ORDER BY day;

Credit Balance Monitoring

# Current credit status
curl -s https://openrouter.ai/api/v1/auth/key \
  -H "Authorization: Bearer $OPENROUTER_API_KEY" | jq '{
    credits_used: .data.usage,
    credit_limit: .data.limit,
    remaining: ((.data.limit // 0) - .data.usage),
    daily_burn_rate: "check analytics DB"
  }'

Weekly Report Generator

def weekly_report(conn) -> str:
    """Generate a text-based weekly analytics report."""
    summary = conn.execute("""
        SELECT COUNT(*) as requests,
               ROUND(SUM(total_cost), 2) as cost,
               ROUND(AVG(latency_ms)) as avg_latency,
               SUM(prompt_tokens + completion_tokens) as tokens
        FROM metrics WHERE timestamp > datetime('now', '-7 days')
    """).fetchone()

    top_models = conn.execute("""
        SELECT model_used, COUNT(*) as n, ROUND(SUM(total_cost), 4) as cost
        FROM metrics WHERE timestamp > datetime('now', '-7 days')
        GROUP BY model_used ORDER BY cost DESC LIMIT 5
    """).fetchall()

    report = f"""
=== OpenRouter Weekly Report ===
Period: Last 7 days
Requests: {summary[0]:,}
Total Cost: ${summary[1]:.2f}
Avg Latency: {summary[2]:.0f}ms
Total Tokens: {summary[3]:,}
Avg Cost/Request: ${summary[1]/max(summary[0],1):.4f}

Top Models by Cost:
"""
    for model, count, cost in top_models:
        report += f"  {model}: {count} requests, ${cost:.4f}\n"

    return report

Output

• A JSON-logged metric per request: timestamp, generationid, modelrequested vs modelused, prompt/completion tokens, exact totalcost, latencyms, and userid
• An openrouter_analytics.db sqlite store with an indexed metrics table ready for the daily/model/user/hourly queries
• A credit-status JSON from the curl + jq snippet: creditsused, creditlimit, and remaining
• A weekly text report with request count, total cost, average latency, total tokens, avg cost/request, and the top 5 models by cost

Examples

Track a single call and read the captured metric:

response, metric = tracked_completion(
    [{"role": "user", "content": "Summarize HTTP/2 in one line"}],
    model="openai/gpt-4o-mini", user_id="alice", max_tokens=60,
)
print(metric["model_used"], metric["total_cost"], metric["latency_ms"])
# openai/gpt-4o-mini 8.4e-05 912.3

After a week of stored metrics, print(weekly_report(conn)) renders the === OpenRouter Weekly Report === block with totals and top models. More worked examples: references/examples.md.

Error Handling

Enterprise Considerations

• Query /api/v1/generation?id= after each request for exact cost (don't estimate from token counts)
• Aggregate raw metrics to hourly/daily summaries after 30 days to manage storage growth
• Build automated weekly reports with cost trends, top users, and anomaly detection
• Set alerts on daily cost exceeding 2x historical average (anomaly detection)
• Track modelrequested vs modelused to monitor fallback frequency
• Use the hourly request pattern to capacity-plan API key rate limits

References

• Examples | Errors
• Generation API | Auth API