llm-cost-autopilot
v0.7.0 · last 7 days
12.3%
cheaper than routing everything to the cloud model ($0.000119 saved on 11 requests)
11
requests
$0.000846
actual cost
$0.000965
baseline cost
10198 ms
avg latency
09-23
09-24
baseline
actual
By model
Model
Requests
Cost
Avg latency
ollama-local
7
$0.000000
15281 ms
groq-20b
4
$0.000846
1304 ms