Optimization Overview

Monitor your LLM spend, savings, and optimization decision activity

SDK Interceptor Live
Last synced: Just now
Net Financial Impact
$0.00

Saved through intelligent model routing and prompt completion caching.

Baseline Cost: $0.00
Actual Spend: $0.00
Savings Rate: 0.0%
Actual Spend
$0.00
Net API spend via CostOpt LIVE
Efficiency Gain
0.0%
Cost reduction ratio OPTIMIZED
Optimization Opportunities
$0.00
Est. unrealized savings ACTIONABLE
System Health
100%
SQLite WAL • Engine Active HEALTHY

Top Optimization Opportunity

Highest impact FinOps action detected
No optimization opportunities detected
Current request traffic is being optimized effectively.

Cost Variance Anomalies

Unexpected spend variance flags
No active anomalies
Recent spend is within expected statistical range.

Recent Telemetry Feed

Latest 5 intercepted LLM API calls
Timestamp Prompt Hash Requested Model Executed Model Latency Cost Status
No telemetry transactions recorded yet.
Actual LLM Spend
$0.00

Net financial outlay after CostOpt optimization algorithms.

Baseline Cost: $0.00
Total Savings: $0.00
Cost Reduction: 0.0%

Spend by Model

Actual spend distribution across models
No spend data available.

Spend by Provider

Actual spend distribution across providers
No provider data available.

Model Cost Breakdown

Operational usage, cost, and savings breakdown per model
Model Provider Request Count Baseline Cost Actual Cost Savings Savings Rate
No model usage recorded.
Total Net Savings
$0.00
Net savings generated by CostOpt
Cache Savings
$0.00
0 requests served from cache
Reroute Savings
$0.00
0 requests model-rerouted
Optimization Rate
0.0%
0 of 0 optimized

Decision Strategy Distribution

Execution strategy selected by DecisionEngine
Loading decision stats...

Task Classification Breakdown

Request intent classification by RequestAnalyzer
Loading classification stats...

Active Optimization Strategies

Operational status of active intelligence layers
Loading active optimization engines...

Optimization Activity Log

Real-time audit log of local optimization decisions
Timestamp Requested Model Executed Model Optimization Type Cost Before Cost After Savings
No optimization events recorded yet. Send requests through the CostOpt SDK to begin tracking activity.

Unrealized Savings Opportunities

Automated policy recommendations for additional cost reduction
Loading recommendations...
Total Requests
0
Total transactions recorded
Optimized Requests
0
Cached or model-rerouted
Cache Hits
0
Served from local SQLite
Average Latency
0 ms
Average roundtrip response
Showing 0 records

Request Explorer

Click any row or Inspect button to open complete Decision Intelligence Trace
Timestamp Prompt Hash Requested Model Executed Model Outcome Latency Actual Cost Savings Action
No requests matching current filter.
Active Policies
2 Rules
Active routing & caching rules
Prompt Cache Engine
Active
MD5 + Cosine similarity threshold
Model Router
Active
Rule-based intent routing
Config State
Synced
costopt.yaml status

Active Policy Rules

Current cost optimization routing definitions
ACTIVE Rule 1
Simple Text Classification
Keywords: classify, sentiment, yes/no, label (max len: 500)
gpt-4o
➔
gpt-4o-mini / deepseek-r1
ACTIVE Rule 2
Short Summary Translation
Keywords: summarize, translate, tldr (max len: 800)
claude-3-5-sonnet
➔
claude-3-haiku

Policy Configuration File (costopt.yaml)

Edit routing rules and model fallbacks in YAML format

Destructive Operations

Actions in this area reset or wipe stored local state and metrics.