Latitude
Spanish MIT-licensed observability for AI agents that groups failing traces into issues and keeps hosted data in Europe
Quick Overview
| Company | Latitude Data S.L. |
|---|---|
| Category | LLM Observability and Evaluation Tools |
| Headquarters | Barcelona, Spain |
| Founded | Not stated |
| EU Presence | Yes - Spain (EU) |
| Data Location | Hosted data stored and processed on European servers |
| Open Source | MIT |
| GDPR Compliant | Yes |
| Self-Hosting | Enterprise plan lists on-premises deployment; code is open source |
| Main Features | Agent tracing, OpenTelemetry ingestion, issue discovery, failure clustering, custom signals, human annotations, datasets, alerts, MCP server |
| Pricing | Free Starter (20K credits/month); Pro $99/month (100K credits); Enterprise on request |
| Best For | Teams running agents in production that want failures grouped and triaged |
| Replaces | LangSmith, Helicone, Arize |
Detailed Review
Latitude positions itself around what happens after traces arrive. Failed sessions become signals, similar failures are clustered into one issue with a trend and example traces, and alerts go to Slack, email or webhooks. Teams triage patterns instead of reading individual logs.
The service is operated by Latitude Data S.L., a Spanish company registered in Barcelona, whose terms are governed by Spanish law with disputes heard in Barcelona. The site states that hosted data stays in Europe and that the product is SOC 2 Type II certified.
From traces to issues
Ingestion works through the vendor SDK or any OpenTelemetry pipeline, so there is no proprietary trace format. Semantic search, text and metadata filters help find cohorts of sessions.
Human annotations on any trace or span feed datasets, and the vendor describes building golden datasets from validated production traces for regression runs.
Open source and deployment
The code is MIT-licensed on the vendor's GitHub. The pricing page lists custom on-premises and cloud deployment on the Enterprise plan, and a Latitude MCP server lets a coding agent work with the workspace.
Plans
Starter is free with 20K credits a month, 30-day retention and unlimited seats. Pro is $99 a month with 100K credits and 90-day retention, with extra credits at $20 per 10K.
Limitations to Consider
Prices are listed in US dollars and usage is metered in credits, so costs depend on how credits are consumed. Retention is 30 days on the free plan and 90 days on Pro. Evaluation depth is built from production data, so teams that want to generate synthetic test cases first may prefer Galtea. Self-hosting support is tied to the Enterprise plan on the pricing page.
Who Should Use Latitude
Teams with agents in production that want an open-source, European-hosted tool that surfaces failure patterns without manual log reading.
Pros and Cons
Pros
- MIT licence
- Failures clustered into issues with trends
- OpenTelemetry-compatible ingestion
- Hosted data stays in Europe
- Unlimited seats on every plan
Cons
- Credit-based metering
- Pricing shown in US dollars
- Free plan retention is 30 days
- On-premises deployment is an Enterprise feature
Alternatives to Latitude
Other European llm observability and evaluation tools worth comparing before you decide:
LangWatch
Netherlands · Open-source agent tracing, evaluations and simulations from Amsterdam
Galtea
Spain · Test generation, simulation and evaluation for regulated industries
Giskard
France · Red teaming and evaluation for LLM agents before release
Orq.ai
Netherlands · AI gateway with built-in traces, evaluation and governance
See all European llm observability and evaluation tools or compare them directly against LangSmith on our alternatives page.
Frequently Asked Questions
Yes. The vendor's GitHub repository is licensed under MIT, and the website describes the product as open source.
The security section of the site says hosted data is stored and processed on European servers.
Yes. The homepage says you can use the SDK or point an existing OpenTelemetry pipeline at Latitude.