Aviationstack is a scalable REST API for real-time flight tracking, historical flight data, airport info, and global aviation data.
from $49.99/mo
About Spanlens
What is Spanlens?
Spanlens is an open-source (MIT) LLM observability platform that gives you complete visibility into every call to OpenAI, Anthropic, and Gemini. It logs requests in one line of code, tracking cost, latency, tokens, and full request bodies. Unlike heavyweight agents or SDK rewrites, Spanlens works by placing a fast proxy in front of your LLM provider—just change your baseURL and you're done.
Why teams use Spanlens
LLM costs spiral quickly, and debugging multi-step agent workflows is painful. Spanlens solves this with per-request cost breakdowns, daily rollups, budget alerts before you exceed your monthly quota, and waterfall-style tracing for agent workflows. It also flags anomalies (3σ deviations from your 7-day baseline), detects PII and prompt-injection patterns, and recommends cheaper models with dollar-figure savings—like suggesting gpt-4o-mini for classification tasks.
Who is Spanlens for?
Spanlens is built for developers and small-to-mid-size teams shipping LLM-powered features in production. It's equally useful for independent developers who want a lightweight, self-hostable monitoring layer, and for platform teams who need per-team cost visibility and budget governance. If you're just experimenting with one API call a day, the free tier may be overkill—but if you're deploying LLM features used by hundreds of users, Spanlens helps you avoid sticker shock and performance surprises.
Real use cases
Use Spanlens to monitor a customer-support chatbot built on Claude, trace a multi-step LangGraph agent that calls multiple tools, or track per-end-user consumption to find the customer consuming the most tokens. It exports data via CSV/JSON and integrates with Slack for alerting, making it a practical companion for any production LLM stack.
Key features
- Request logging — Every call's model, tokens, cost, latency, and full body, filterable and exportable.
- Cost tracking — Per-request breakdowns, daily rollups, and budget alerts via Slack or webhook.
- Agent tracing — Multi-step workflows visualized as waterfall span trees; spot the single step that takes 18s.
- Anomaly detection — Flags 3σ deviations in latency or cost vs. your 7-day baseline.
- PII and injection scan — Regex detection on request bodies; API keys auto-masked before storage.
- Model recommender — Suggests cheaper models for your workloads with dollar-figure savings.
- Evals — LLM-as-judge scores every response 0 to 1 to compare versions.
SaaSpartout Score
7.7
/10
Editorial score from our review methodology — not user ratings.
◆ AI advisor — 30 seconds, no signup
Why are you looking at Monitoring tools today?
Spanlens Pricing
Spanlens pricing: from $0/mo (free) · Pro $29/mo. Billing model: Freemium.
For comparison: the median starting price in Developer Tools is $19.99/month, measured across 301 tools we track. See the full SaaS Pricing Index →
Free
$0/mo — 50K requests/month, 1 seat, 1 workspace, unlimited projects, 14-day log retention, all core features, CSV/JSON export, community support.
Pro (Most popular)
$29/mo + $8 per extra 100K requests — 100K requests/month, 3 seats, 2 workspaces, unlimited projects, 90-day log retention, 5 alerts, email notifications, email support.
Team
$149/mo + $5 per extra 100K requests — 1M requests/month, 10 seats, 5 workspaces, unlimited projects, 365-day log retention, unlimited alerts, Slack notifications, webhooks, priority support.
Enterprise
Custom — custom requests/month, unlimited seats/workspaces/projects, 365-day log retention, unlimited alerts, SSO (SAML/Okta), dedicated support + SLA. Self-hostable for free with MIT license (no ee/ folder).
All paid plans include overage billing with a soft limit, hard cap (default 5×), and cost-certainty mode. Prices in USD, exclusive of taxes.
Frequently asked questions
How much does Spanlens cost?
Spanlens offers a free plan with 50K requests/month, a Pro plan at $29/mo (100K requests), Team at $149/mo (1M requests), and custom Enterprise pricing. You can also self-host it for free via Docker.
Is Spanlens free?
Yes, there's a free plan for personal projects (50K requests/month, 14-day log retention). The entire platform is MIT open-source, so you can self-host all features at no cost.
What is Spanlens used for?
Spanlens is an LLM observability platform that logs every API call to providers like OpenAI, Anthropic, and Gemini, tracking cost, latency, tokens, and errors. It helps developers monitor and optimize LLM usage in production.
How does Spanlens work?
You install the SDK (npm or pip) and change your baseURL to point to Spanlens. It acts as a proxy that logs each request, analyzes it for anomalies and PII, and provides dashboards for cost and latency.
What are the best Spanlens alternatives?
Top alternatives include Langfuse Cloud (starts at $59/mo) and Helicone (Pro at $50/mo). Spanlens is cheaper and fully self-hostable, making it a strong choice for developers on a budget.
Who is Spanlens best for?
Spanlens is best for developers and small-to-mid-size teams shipping LLM features in production, especially those needing granular cost tracking and agent tracing. It's not ideal for hobbyists who only make occasional API calls.
