Account Usage Analytics

Track your TensorOne account usage including compute hours, costs, GPU utilization, and resource consumption. Get detailed analytics to optimize your AI workloads and manage costs effectively.

Get Usage Summary

Retrieve overall usage statistics for your account.

curl -X GET "https://api.tensorone.ai/v2/accounts/usage" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -G -d "period=30d"

Query Parameters

period: Time period (1d, 7d, 30d, 90d, 180d)
timezone: Timezone for date calculations (default: account timezone)

Response

{
    "period": "30d",
    "periodStart": "2024-02-20T00:00:00Z",
    "periodEnd": "2024-03-20T23:59:59Z",
    "summary": {
        "totalCost": 1250.75,
        "totalComputeHours": 485.2,
        "totalEndpointExecutions": 12450,
        "totalDataTransfer": "125.8GB",
        "averageDailyCost": 41.69,
        "costTrend": "+12.5%"
    },
    "breakdown": {
        "compute": {
            "cost": 980.5,
            "hours": 485.2,
            "percentage": 78.4
        },
        "storage": {
            "cost": 125.25,
            "gb": "2.5TB",
            "percentage": 10.0
        },
        "dataTransfer": {
            "cost": 95.0,
            "gb": "125.8GB",
            "percentage": 7.6
        },
        "other": {
            "cost": 50.0,
            "percentage": 4.0
        }
    },
    "topResources": [
        {
            "type": "gpu",
            "name": "NVIDIA A100",
            "cost": 645.3,
            "hours": 201.5,
            "percentage": 51.6
        },
        {
            "type": "gpu",
            "name": "NVIDIA RTX 4090",
            "cost": 335.2,
            "hours": 283.7,
            "percentage": 26.8
        }
    ]
}

Get Detailed Usage

Retrieve detailed usage data with timeline breakdown.

curl -X GET "https://api.tensorone.ai/v2/accounts/usage/detailed" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -G -d "period=7d" -d "granularity=day"

Query Parameters

period: Time period (1d, 7d, 30d, 90d)
granularity: Data granularity (hour, day, week)
resources: Filter by resource types (gpu, storage, transfer)

Response

{
    "period": "7d",
    "granularity": "day",
    "timeline": [
        {
            "date": "2024-03-14",
            "cost": 45.2,
            "computeHours": 18.5,
            "executions": 420,
            "dataTransfer": "4.2GB",
            "resources": {
                "gpu": {
                    "a100": { "hours": 12.0, "cost": 38.4 },
                    "rtx4090": { "hours": 6.5, "cost": 6.8 }
                }
            }
        },
        {
            "date": "2024-03-15",
            "cost": 52.8,
            "computeHours": 21.2,
            "executions": 485,
            "dataTransfer": "5.1GB",
            "resources": {
                "gpu": {
                    "a100": { "hours": 15.5, "cost": 49.6 },
                    "rtx4090": { "hours": 5.7, "cost": 3.2 }
                }
            }
        }
    ],
    "projections": {
        "monthlyEstimate": 1650.0,
        "dailyAverage": 48.75,
        "confidence": 0.85
    }
}

Get GPU Usage

Get detailed GPU usage statistics and performance metrics.

curl -X GET "https://api.tensorone.ai/v2/accounts/usage/gpu" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -G -d "period=30d" -d "includeMetrics=true"

Response

{
    "period": "30d",
    "gpuUsage": [
        {
            "gpuType": "NVIDIA A100",
            "totalHours": 201.5,
            "totalCost": 645.3,
            "averageUtilization": 87.2,
            "peakUtilization": 98.5,
            "sessions": 145,
            "averageSessionDuration": "1h 23m",
            "performance": {
                "averageMemoryUsage": 75.8,
                "averageTemperature": 68.5,
                "efficiency": 0.92
            },
            "usage": [
                {
                    "date": "2024-03-01",
                    "hours": 8.5,
                    "cost": 27.2,
                    "utilization": 89.2,
                    "sessions": 6
                }
            ]
        },
        {
            "gpuType": "NVIDIA RTX 4090",
            "totalHours": 283.7,
            "totalCost": 335.2,
            "averageUtilization": 82.4,
            "peakUtilization": 95.1,
            "sessions": 201,
            "averageSessionDuration": "1h 25m",
            "performance": {
                "averageMemoryUsage": 68.5,
                "averageTemperature": 71.2,
                "efficiency": 0.88
            }
        }
    ]
}

Get Endpoint Usage

Retrieve usage statistics for serverless endpoints.

curl -X GET "https://api.tensorone.ai/v2/accounts/usage/endpoints" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -G -d "period=30d"

Response

{
    "period": "30d",
    "endpointUsage": [
        {
            "endpointId": "ep_1234567890abcdef",
            "name": "stable-diffusion-xl",
            "executions": 5420,
            "computeHours": 125.8,
            "cost": 402.56,
            "averageExecutionTime": "1.4s",
            "successRate": 98.7,
            "usage": [
                {
                    "date": "2024-03-01",
                    "executions": 180,
                    "computeHours": 4.2,
                    "cost": 13.44,
                    "averageExecutionTime": "1.3s"
                }
            ],
            "topUsers": [
                {
                    "userId": "usr_abc123",
                    "executions": 1250,
                    "percentage": 23.1
                }
            ]
        }
    ],
    "summary": {
        "totalExecutions": 12450,
        "totalCost": 980.5,
        "averageExecutionCost": 0.0788,
        "mostUsedEndpoint": "stable-diffusion-xl"
    }
}

Get Cost Breakdown

Get detailed cost breakdown by service, resource type, and time period.

curl -X GET "https://api.tensorone.ai/v2/accounts/usage/costs" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -G -d "period=30d" -d "breakdown=service"

Query Parameters

breakdown: Breakdown type (service, resource, region, team)
currency: Currency code (default: USD)

Response

{
    "period": "30d",
    "currency": "USD",
    "breakdown": "service",
    "totalCost": 1250.75,
    "services": [
        {
            "service": "GPU Compute",
            "cost": 980.5,
            "percentage": 78.4,
            "details": {
                "a100": { "cost": 645.3, "hours": 201.5 },
                "rtx4090": { "cost": 335.2, "hours": 283.7 }
            }
        },
        {
            "service": "Storage",
            "cost": 125.25,
            "percentage": 10.0,
            "details": {
                "model_storage": { "cost": 75.0, "gb": 1500 },
                "data_storage": { "cost": 50.25, "gb": 1005 }
            }
        },
        {
            "service": "Data Transfer",
            "cost": 95.0,
            "percentage": 7.6,
            "details": {
                "ingress": { "cost": 0.0, "gb": 45.2 },
                "egress": { "cost": 95.0, "gb": 125.8 }
            }
        },
        {
            "service": "API Requests",
            "cost": 50.0,
            "percentage": 4.0,
            "details": {
                "management_api": { "cost": 25.0, "requests": 125000 },
                "inference_api": { "cost": 25.0, "requests": 12450 }
            }
        }
    ],
    "trends": {
        "weekOverWeek": "+8.5%",
        "monthOverMonth": "+12.3%"
    }
}

Get Usage Alerts

Configure and retrieve usage alerts and notifications.

curl -X GET "https://api.tensorone.ai/v2/accounts/usage/alerts" \
  -H "Authorization: Bearer YOUR_API_KEY"

Response

{
    "alerts": [
        {
            "id": "alert_1234567890",
            "name": "Monthly Cost Alert",
            "type": "cost",
            "threshold": 1000.0,
            "current": 1250.75,
            "percentage": 125.1,
            "status": "triggered",
            "triggeredAt": "2024-03-18T14:30:00Z",
            "notificationsSent": 2
        },
        {
            "id": "alert_0987654321",
            "name": "GPU Hours Alert",
            "type": "compute_hours",
            "threshold": 400.0,
            "current": 485.2,
            "percentage": 121.3,
            "status": "triggered",
            "triggeredAt": "2024-03-19T09:15:00Z",
            "notificationsSent": 1
        }
    ],
    "settings": {
        "emailNotifications": true,
        "slackNotifications": true,
        "webhookUrl": "https://yourapp.com/webhooks/usage-alerts"
    }
}

Code Examples

Python SDK

from tensorone import TensorOneAPI
from datetime import datetime, timedelta

client = TensorOneAPI(api_key="your_api_key")

# Get usage summary
usage = client.accounts.usage.get_summary(period="30d")
print(f"Total cost: ${usage.summary.total_cost}")
print(f"Compute hours: {usage.summary.total_compute_hours}")

# Get detailed timeline
detailed = client.accounts.usage.get_detailed(
    period="7d",
    granularity="day"
)

for day in detailed.timeline:
    print(f"{day.date}: ${day.cost} ({day.compute_hours}h)")

# Get GPU usage
gpu_usage = client.accounts.usage.get_gpu(
    period="30d",
    include_metrics=True
)

for gpu in gpu_usage.gpu_usage:
    print(f"{gpu.gpu_type}: {gpu.total_hours}h @ {gpu.average_utilization}% util")

# Get cost breakdown
costs = client.accounts.usage.get_costs(
    period="30d",
    breakdown="service"
)

for service in costs.services:
    print(f"{service.service}: ${service.cost} ({service.percentage}%)")

# Set up cost alert
alert = client.accounts.usage.create_alert(
    name="Production Cost Alert",
    type="cost",
    threshold=2000.00,
    period="monthly"
)

# Export usage data for analysis
usage_data = client.accounts.usage.export(
    period="90d",
    format="csv",
    include_details=True
)

JavaScript SDK

import { TensorOneAPI } from "@tensorone/sdk";

const client = new TensorOneAPI("your_api_key");

// Get usage summary
const usage = await client.accounts.usage.getSummary({ period: "30d" });
console.log(`Total cost: $${usage.summary.totalCost}`);
console.log(`Compute hours: ${usage.summary.totalComputeHours}`);

// Get detailed usage with timeline
const detailed = await client.accounts.usage.getDetailed({
    period: "7d",
    granularity: "day",
});

detailed.timeline.forEach((day) => {
    console.log(`${day.date}: $${day.cost} (${day.computeHours}h)`);
});

// Get GPU usage statistics
const gpuUsage = await client.accounts.usage.getGpu({
    period: "30d",
    includeMetrics: true,
});

gpuUsage.gpuUsage.forEach((gpu) => {
    console.log(`${gpu.gpuType}: ${gpu.totalHours}h @ ${gpu.averageUtilization}% util`);
});

// Get endpoint usage
const endpoints = await client.accounts.usage.getEndpoints({ period: "30d" });
endpoints.endpointUsage.forEach((endpoint) => {
    console.log(`${endpoint.name}: ${endpoint.executions} executions, $${endpoint.cost}`);
});

// Create usage alert
const alert = await client.accounts.usage.createAlert({
    name: "Development Cost Alert",
    type: "cost",
    threshold: 500.0,
    period: "monthly",
});

// Generate usage report
const report = await client.accounts.usage.generateReport({
    period: "30d",
    format: "pdf",
    includeCharts: true,
});

Error Responses

400 Bad Request

{
    "error": "INVALID_PERIOD",
    "message": "Invalid time period specified",
    "details": {
        "validPeriods": ["1d", "7d", "30d", "90d", "1y"],
        "received": "invalid_period"
    }
}

403 Forbidden

{
    "error": "INSUFFICIENT_PERMISSIONS",
    "message": "Access to billing information requires billing scope",
    "details": {
        "requiredScope": "billing"
    }
}

429 Too Many Requests

{
    "error": "RATE_LIMIT_EXCEEDED",
    "message": "Usage API rate limit exceeded",
    "details": {
        "limit": 100,
        "period": "1h",
        "resetAt": "2024-03-20T18:00:00Z"
    }
}

Best Practices

Cost Optimization

Monitor usage trends to identify cost spikes
Set up proactive alerts before reaching budget limits
Use detailed breakdowns to identify optimization opportunities
Track GPU utilization to right-size resources

Resource Management

Analyze endpoint performance to optimize execution times
Monitor storage usage and clean up unused data
Track data transfer costs and optimize data flow
Use projections for capacity planning

Monitoring and Alerts

Set multiple alert thresholds (warning and critical)
Configure alerts for different metrics (cost, hours, executions)
Use webhooks for real-time notifications
Regular review of usage patterns and trends

Use Cases

Budget Management

# Set up comprehensive budget monitoring
def setup_budget_alerts():
    # Monthly cost alert at 80% of budget
    client.accounts.usage.create_alert(
        name="Monthly Budget Warning",
        type="cost",
        threshold=4000.00,  # 80% of $5000 budget
        period="monthly"
    )

    # Daily cost alert for unexpected spikes
    client.accounts.usage.create_alert(
        name="Daily Cost Spike",
        type="cost",
        threshold=200.00,
        period="daily"
    )

Performance Optimization

# Analyze GPU efficiency
def analyze_gpu_efficiency():
    gpu_usage = client.accounts.usage.get_gpu(period="30d", include_metrics=True)

    for gpu in gpu_usage.gpu_usage:
        efficiency = gpu.performance.efficiency
        utilization = gpu.average_utilization

        if efficiency < 0.8 or utilization < 70:
            print(f"GPU {gpu.gpu_type} may need optimization:")
            print(f"  Efficiency: {efficiency:.2f}")
            print(f"  Utilization: {utilization:.1f}%")

Cost Attribution

# Track costs by team or project
def generate_team_report():
    costs = client.accounts.usage.get_costs(
        period="30d",
        breakdown="team"
    )

    for team in costs.teams:
        print(f"Team {team.name}: ${team.cost}")
        # Generate detailed team report
        team_report = client.accounts.usage.get_detailed(
            period="30d",
            filters={"team_id": team.id}
        )

Capacity Planning

# Use projections for capacity planning
def plan_capacity():
    detailed = client.accounts.usage.get_detailed(period="90d")
    monthly_estimate = detailed.projections.monthly_estimate

    if monthly_estimate > 8000:  # Budget threshold
        print("Warning: Projected monthly cost exceeds budget")
        print(f"Estimated: ${monthly_estimate}")

        # Analyze growth trend
        print(f"Confidence: {detailed.projections.confidence:.1%}")

Getting Started

Account Management

GPU Clusters (VPS)

Serverless Endpoints

Managed Training

AI Services

Payment & Billing

Monitoring & Analytics

Account Usage Analytics

Get Usage Summary

Query Parameters

Response

Get Detailed Usage

Query Parameters

Response

Get GPU Usage

Response

Get Endpoint Usage

Response

Get Cost Breakdown

Query Parameters

Response

Get Usage Alerts

Response

Code Examples

Python SDK

JavaScript SDK

Error Responses

400 Bad Request

403 Forbidden

429 Too Many Requests

Best Practices

Cost Optimization

Resource Management

Monitoring and Alerts

Use Cases

Budget Management

Performance Optimization

Cost Attribution

Capacity Planning

Getting Started

Account Management

GPU Clusters (VPS)

Serverless Endpoints

Managed Training

AI Services

Payment & Billing

Monitoring & Analytics

​Get Usage Summary

​Query Parameters

​Response

​Get Detailed Usage

​Query Parameters

​Response

​Get GPU Usage

​Response

​Get Endpoint Usage

​Response

​Get Cost Breakdown

​Query Parameters

​Response

​Get Usage Alerts

​Response

​Code Examples

​Python SDK

​JavaScript SDK

​Error Responses

​400 Bad Request

​403 Forbidden

​429 Too Many Requests

​Best Practices

​Cost Optimization

​Resource Management

​Monitoring and Alerts

​Use Cases

​Budget Management

​Performance Optimization

​Cost Attribution

​Capacity Planning

Get Usage Summary

Query Parameters

Response

Get Detailed Usage

Query Parameters

Response

Get GPU Usage

Response

Get Endpoint Usage

Response

Get Cost Breakdown

Query Parameters

Response

Get Usage Alerts

Response

Code Examples

Python SDK

JavaScript SDK

Error Responses

400 Bad Request

403 Forbidden

429 Too Many Requests

Best Practices

Cost Optimization

Resource Management

Monitoring and Alerts

Use Cases

Budget Management

Performance Optimization

Cost Attribution

Capacity Planning