Total Cost (MTD)
$6,742.00
+12.3%vs last month
Projected Monthly
$8,920.45
at current rate
Total Savings
$1,345.67
16.6% of spend
Optimization Score
78/100
room for improvement

Provider Breakdown

OpenAI$2,847.32
42.3%45,892 req
Anthropic$1,956.78
29.1%12,453 req
Google$1,023.45
15.2%28,456 req
Cohere$512.89
7.6%8,934 req
Mistral AI$302.12
4.5%5,621 req
Together AI$99.44
1.5%2,845 req

Optimization Suggestions

6 available
Switch GPT-4 to GPT-4o
HIGH

GPT-4o is 50% cheaper with faster response times

Save $892.45/mo
Enable Claude 3.5 caching
HIGH

Use cached tokens for repeated contexts (40% discount)

Save $456.32/mo
Batch API for embeddings
MEDIUM

Process embeddings in batches for 70% cost reduction

Save $312.67/mo
Switch to Gemini 1.5 Flash
MEDIUM

Flash model is 95% cheaper for non-critical tasks

Save $234.89/mo
Add retry logic for failures
LOW

Implement exponential backoff to reduce wasted requests

Save $178.23/mo
Use Mistral for simple tasks
LOW

Switch non-complex prompts to Mistral Small

Save $145.67/mo
Date Provider ModelRequests TokensCost
2026-02-24OpenAIgpt-4o1,234630,000$45.67
2026-02-24Anthropicclaude-3-5-sonnet456252,000$32.45
2026-02-23OpenAIgpt-41,567730,000$89.23
2026-02-23Googlegemini-1.5-pro892323,000$28.45
2026-02-22Anthropicclaude-3-opus234201,000$56.78
2026-02-22OpenAIgpt-4-turbo1,456675,000$67.89
2026-02-21Coherecommand-r-plus678245,000$23.45
2026-02-21Mistrallarge345123,000$18.92