Tokentra
Integrations

Connect Your Entire AI Stack

TokenTra integrates with all major AI providers and notification channels. Connect in minutes, get visibility instantly.

AI Providers

Connect all your AI providers for unified cost visibility

OpenAI

Available

Connect your OpenAI organization to track costs across all models including GPT-5, o3, DALL-E 4, Whisper, and Embeddings.

Supported Models

  • GPT-5, GPT-5.1, GPT-5 Pro
  • o3, o3-mini (advanced reasoning models)
  • GPT-4o, GPT-4o-mini (legacy)
  • DALL-E 4
  • Whisper v3 (speech-to-text)
  • TTS HD (text-to-speech)
  • text-embedding-4-small, text-embedding-4-large

Data Available

  • Token usage by model and date
  • Cost in USD
  • Request counts
  • 1-minute granularity
Auth: Admin API Key (read-only access to usage data)
Setup: ~1 minute

Anthropic

Available

Track your Claude usage and costs across all model variants including the latest Claude Opus 4.5.

Supported Models

  • Claude Opus 4.5 (flagship)
  • Claude Sonnet 4
  • Claude Haiku 4
  • Claude 3.5 Sonnet (legacy)
  • Claude 3.5 Haiku (legacy)

Data Available

  • Input/output tokens
  • Cached tokens and cache creation tokens
  • Cost breakdown by model and workspace
  • 1-minute, 1-hour, or 1-day granularity
Auth: Admin API Key (sk-ant-admin-...)
Setup: ~1 minute

Google Vertex AI

Available

Connect your Google Cloud Vertex AI usage for Gemini 3 and other models.

Supported Models

  • Gemini 3 Pro, Gemini 3 Flash, Gemini 3 Ultra
  • Gemini 2 Pro Vision (legacy)
  • Imagen 3 (image generation)
  • Embeddings v2

Data Available

  • Token usage via BigQuery export
  • Cost via Cloud Billing API
  • Per-model breakdown
Auth: Service account with BigQuery billing export access
Setup: ~5 minutes

Azure OpenAI

Available

Connect your Azure OpenAI deployments to TokenTra with full support for GPT-5 and o3 models.

Supported Models

  • GPT-5, GPT-5.1, GPT-5 Pro via Azure
  • o3, o3-mini via Azure
  • GPT-4o (legacy)
  • DALL-E 4
  • Embeddings v4

Data Available

  • Cost data via Azure Cost Management API
  • Usage metrics via Azure Monitor
  • Per-deployment breakdown
Auth: Service principal authentication
Setup: ~5 minutes

AWS Bedrock

Available

Full support for AWS Bedrock as a first-class provider with all major foundation models.

Supported Models

  • Claude Opus 4.5, Claude Sonnet 4 (via Bedrock)
  • Amazon Titan v2
  • Llama 3.2, Llama 3.1
  • Mistral Large 2
  • Stable Diffusion 3

Data Available

  • Full cost and usage sync via AWS Cost Explorer API
  • Unified dashboard view alongside other providers
Auth: IAM role-based authentication (cross-account)
Setup: ~5 minutes

Notification Channels

Get alerts where your team works

Slack

Available

Get alerts where your team already works.

  • Real-time alert notifications
  • Weekly digest summaries
  • Slash commands for quick lookups
  • Interactive budget approvals

Email

Available

Receive alerts and reports directly in your inbox.

  • Instant alert notifications
  • Daily/weekly digest options
  • PDF report attachments
  • Customizable templates

PagerDuty

Available

Integrate with your incident management workflow.

  • Critical alert escalation
  • On-call routing
  • Incident creation
  • Auto-resolve on budget recovery

Webhooks

Available

Send alerts to any endpoint for custom integrations.

  • JSON payload delivery
  • Retry logic with exponential backoff
  • HMAC signature verification
  • Custom headers support

Coming Soon

More integrations on the way

Microsoft Teams

Native Teams integration for alerts and reports

Coming Soon

Datadog

Send metrics to Datadog for unified observability

Coming Soon

Jira

Create tickets automatically for budget overruns

Coming Soon

Linear

Track cost optimization tasks in Linear

Coming Soon

Ready to Connect Your Providers?

Get started in minutes. Connect your first provider and see your AI costs instantly.