Hanzo Console
LLM observability and tracing
Track every LLM call across your stack. Monitor costs, latency, quality, and token usage. Debug production issues with full trace visibility.
Full Visibility Into Your LLM Stack
Debug, optimize, and monitor every AI interaction in production
Trace Explorer
Drill into any LLM call. See prompts, completions, token counts, and latencies with full nested span trees.
Cost Tracking
Real-time cost monitoring by model, user, feature, and environment. Set budgets and alerts.
Latency Analysis
P50, P95, P99 latency breakdowns. Identify slow calls, timeouts, and retry storms.
Quality Scores
Automated evaluation with custom scoring functions. Track hallucination rates and relevance.
Prompt Management
Version control prompts. A/B test variations. Roll back to previous versions instantly.
Multi-Tenant
Organization-scoped data isolation. SSO via Hanzo IAM. Role-based access controls.
Integrate in Minutes
Drop-in SDK for Python, TypeScript, and any OpenAI-compatible client
from hanzo import Hanzo
client = Hanzo()
# Every call is automatically traced
response = client.chat.completions.create(
model="zen-480b-instruct",
messages=[{"role": "user", "content": "Hello"}],
metadata={
"user_id": "usr_123",
"session_id": "sess_abc",
}
)
# View traces at console.hanzo.ai25% of compute goes back to open source
Every deployment is SBOM-verified. Contributors to Console earn a share of compute revenue — transparent, on-chain, and customizable by the community.
See What Your LLMs Are Doing
Start tracing in under 5 minutes. Free tier includes 50K traces/month.