Optimization
Stop paying for duplicate LLM calls
Fluiq profiles your real trace history, provisions a cache, and serves repeated prompts automatically.
Fluiq/ optimize1h6h24h7d
Hit Rate
84.3%
10.5k hits saved
Total Calls
12.4k
last 24h
Misses
1.9k
15.7% miss rate
Cache Performance
Overall hit rate84.3%
EmbeddingCache91.2%
PromptCache77.5%
What you get
Server-side
Zero infra to manage
A dedicated cache instance is provisioned for your account. Nothing to deploy or operate.
Profiled
Built from real traffic
The cache profile comes from your actual traces, not a generic one-size heuristic.
Tunable
TTL and model scope
Set time-to-live and which models are eligible. Observe mode measures savings before you commit.
Measure first. Cache what repeats.
- Observe mode records would-be hits without intercepting calls
- Cache mode serves duplicates from a dedicated instance
- Per-model scope and configurable TTL
app.py
import fluiq
fluiq.instrument(api_key="fl_...")
# "observe" measures savings; "cache" serves duplicates
fluiq.optimize(mode="cache")
# Repeated prompts now resolve from your cache
answer(question)Part of the Fluiq platform
Compare plansObservabilityTrace every call, cost, and latency.ExploreSecurityBlock attacks, redact PII and secrets.ExploreEvaluationScore responses and whole agent runs.ExploreDatasetsGolden sets that capture whole agent runs.ExplorePrompt ManagementVersion and deploy prompt templates.ExploreAlertsPush eval and security events to Slack.Explore
Unlimited traces, always free.
Start on the free tier and turn on each pillar as your pipeline grows. No code changes required.