AI Observability Dashboard

Real-time monitoring and evaluation of AI systems

Req/min
127

+5.2%

Avg Latency
1.24s

-0.1%

Error %
2.1%

+0.3%

Token/min
15,420

+8.7%

$/hour
$12.45

+2.1%

Total Requests
24,847

+12% from last hour

Active Models
23

+2 new this week

Avg Quality Score
8.7/10

+0.3 from yesterday

Live Cost
$12.45/hr

+$2.10 from avg

Live Activity Feed
14:32:15
GPT-4 Tool Call

Function: get_weather, Duration: 1.2s

14:32:14
Processing MCP

Context server responding...

14:32:13
API Timeout

External service timeout after 30s

14:32:12
Code Execution

Python script completed successfully

14:32:11
LLM Response

Claude-3: 245 tokens generated

Performance Trends
Evaluation scores over the last 10 weeks
System Health
Real-time platform performance indicators
API Response Timehealthy
Average: 1.2s
Model Availabilityhealthy
21/23 models online
Queue Processingwarning
12 jobs pending
Error Ratehealthy
0.8% last 24h
Recent Notifications
System alerts and important updates

Model Connection Failed

GPT-4 Turbo connection timeout after 3 retry attempts

5 minutes ago

Evaluation Completed

Safety evaluation suite finished with 94.2% pass rate

1 hour ago

High Queue Volume

15 evaluations pending - estimated wait time 45 minutes

2 hours ago

New Model Available

Claude-3.5 Sonnet has been added to your available models

1 day ago
Quick Actions
Common monitoring and evaluation tasks
Recent Evaluations
Latest evaluation runs and their status
GPT-4 Safety Evaluationgpt-4-turbo
94.2%
Score
completed2 hours ago
Claude Performance Testclaude-3-sonnet
runningRunning...
Llama2 Benchmark Suitellama-2-70b
failed1 day ago
Custom Model Validationcustom-model-v2
78.9%
Score
completed2 days ago