Full trace visibility across every LLM call
Every token, latency, and dollar attributed to the exact agent node that spent it. Unlimited traces on every plan; retention is the only paid axis.
Fluiq/ traces| FUNCTION | MODEL | LATENCY | COST | SOURCE |
|---|---|---|---|---|
| answer_question | gpt-4o | 1,243ms | $0.012 | LangChain |
| search_docs | claude-3.5-s | ⚡ cached | $0.000 | Cached |
| generate_report | gpt-4o | 2,108ms | $0.041 | OpenAI |
| classify_intent | gemini-1.5 | 890ms | $0.005 | |
| answer_question | gpt-4o | 1,540ms | $0.019 | LangChain |
What you get
Attribution
Per-node token and cost
Token counts and USD cost at live provider rates, attributed down to the span that spent them.
Latency
p50 / p95 / p99
Latency histograms per agent and per model, so you watch the tail, not just the average.
Multi-agent
Real DAGs, not flat lists
Fan-outs, joins, and loop-backs across LangGraph, CrewAI, and Google ADK render as the graph your agents actually executed.
Live
Real-time streaming
Traces land on the dashboard as calls complete. Watch a run unfold instead of refreshing.
Migration
Bring your history with you
Import existing traces from LangSmith, Langfuse, Phoenix, or Braintrust, so switching does not mean starting from an empty dashboard.
One decorator turns any call into a span.
- Patches OpenAI, Anthropic, Gemini, LangChain, and vector stores automatically
- Nested calls build a full span tree with parent and child timing
- Cost is computed at live provider rates, per model
import fluiq
fluiq.instrument(api_key="fl_...")
@fluiq.trace
def answer(question: str) -> str:
docs = store.search(question, k=5)
return llm.invoke(prompt(question, docs))
# Traced with cost, latency, and a full span treePart of the Fluiq platform
Compare plansUnlimited traces, always free.
Tracing, security, evaluation, datasets, and prompts all run on the free tier. Caching and Slack alerts unlock on a paid plan.