Granular usage insights

Per-Model/Provider Breakdown

Break down usage and spend by provider and model so you can quickly spot expensive outliers.

Get startedTry in Lounge
Requests by provider
  1. 1

    Anthropic

    48%
  2. 2

    OpenAI

    32%
  3. 3

    Google

    15%
  4. 4

    Others

    5%
01

Model-Level Analytics

Track usage and costs for each individual model

02

Provider Comparison

Compare costs and performance across providers

03

Outlier Detection

Quickly identify expensive requests or unusual patterns

04

Optimization Insights

Get recommendations for cost optimization

See it in the dashboard

Model Usage Breakdown
Detailed performance and cost metrics by model
ModelProviderRequestsTokensCostAvg Latency
anthropic/claude-3-5-sonnet-20241022Anthropic4,532456,789$52.181123ms
openai/gpt-4oOpenAI3,421387,654$45.23987ms
google-ai-studio/gemini-1.5-proGoogle1,876198,765$18.451456ms
together/mixtral-8x7bTogether AI2,134245,678$11.57654ms

Total Models

4

Total Requests

11,963

Total Cost

$127.43

Avg Latency

1055ms

Built for

  • Cost Optimization

    Identify and switch from expensive models to cheaper ones

  • Performance Analysis

    Compare model performance for your specific use case

  • Budget Planning

    Forecast future costs based on historical usage

More in observability & cost

All features

Source Attribution

Get credit for your traffic

Data Retention

Keep metadata, or keep everything

Data Streams

Logs to your SIEM and warehouse

Start using Per-Model/Provider Breakdown today

One API key. Every model. Live in minutes.

Start buildingBrowse models