tokevenKnowledge Base
Everything you need to get tracking
Getting Started
Quick start guide
Install the wrapper and start tracking in under two minutes
Installing the Python wrapper
Detailed installation guide for the Tokeven Python SDK
Your first tracked call
Make your first API call through the wrapper and see it in the dashboard
Understanding your dashboard
A tour of the three dashboard views - Spend, Save, Proof - and the pages around them
Dashboards & Analytics
Cost Overview dashboard
Understanding your cost data and spend trends
Model Usage breakdown
See which models your team uses and where to optimize
Token Metrics explained
Understanding token counts, cache rates, and prompt efficiency
Usage Patterns & heatmaps
When and how your team uses AI
Exporting data (CSV, JSON, API)
Get your data out of Tokeven in any format
SDK & Integration
Privacy & Security
Privacy architecture
How Tokeven keeps your prompts private by design
What data leaves your device
A detailed look at the data flow from your machine to Tokeven
Ingest token management
Create, rotate, and revoke your write-only ingest tokens
Optimization & Savings
How pre-query pricing works
See the cost before you run
Model-tier recommendations
How Tokeven suggests the right model for each task
Cache efficiency tips
Maximize cache hits to cut input costs by 90%
How savings are verified
How we measure and prove the dollars you save
Account & Billing
Team management
Add team members, set roles, and manage access
Understanding gainshare pricing
How our 10% of verified savings model works
API rate limits
The ingest and API rate limits we actually enforce, and the one cap that is plan-dependent
Use Cases
Enterprise AI pipelines and internal models
How companies running chatbots, knowledge hubs, and internal AI tools can cut costs with Tokeven
Reducing chatbot costs without losing quality
A practical guide to cutting AI chatbot spend by 40-60% with model tiering and cache optimization