tools
Opik
Opik is an open-source platform for tracing, evaluating, testing, and monitoring LLM applications and AI agents.
In inglese
Opik records LLM calls, tool invocations, agent steps, errors, token usage, and costs. Teams use it to debug applications, evaluate outputs with automated metrics, run test suites, manage prompts, and monitor production systems.
It supports hosted cloud plans and self-hosting. The free plans include core observability and evaluation features, while higher tiers add increased span limits, customizable retention, enterprise deployments, SSO, compliance, and dedicated support.
Features
- Trace LLM calls, tool invocations, agent steps, errors, tokens, and costs
- Evaluate applications with LLM-as-a-judge and heuristic metrics
- Create test suites with item-level and suite-level assertions
- Monitor production traces with online evaluation rules and alerts
- Manage prompt versions and compare prompts in the Prompt Playground
- Optimize prompts with six optimization algorithms
- Self-host with Docker or Kubernetes
- Open-source project; license not read
Use cases
- Debug multi-step agents and retrieval-augmented generation pipelines
- Evaluate chatbot answers for relevance, hallucination, and task completion
- Run regression tests before deploying new prompts or models
- Monitor production quality, latency, errors, token usage, and model cost
- Review and annotate traces with subject-matter experts
- Track coding-agent usage and spending across engineering teams
Pros
Cons
Latest updates
- 2.2.81-6780 (2.2.81-6780)
Provision the extended free-form SQL ClickHouse user.
- 2.2.80 (2.2.80)
- 2.2.79 (2.2.79)
Capabilities
- Agent that edits files — “Opik’s powerful coding assistant analyzes your traces, suggests fixes, and implements them in your development code — with built-in version control and regression testing.” source
- Knows the whole codebase — “Configure it to let Ollie read your agent codebase and write fixes directly to it.” source
- Choice of models — “If you’re not ready to log traces yet, we recommend trying out the prompt playground to experiment with user prompts and system prompts and compare models side by side, with no setup steps required.” source
- API — “Opik’s REST API and complete Python client can be used with both the Open-Source platform and Opik Cloud” source
- Official SDKs — “SDKs (Python, TypeScript)” source
- Traces and evaluates — “Opik logs every step your agent takes, from user interactions to context retrieval and tool calls — with automated eval workflows to find and fix errors across development, testing, and production.” source
Get it
Security
Pricing
- Starting price
- $19/mo
- Prices checked
- 2026-09-24
Open Source
Free
- Full AI observability & agent testing feature set
- True OSS: same codebase as the hosted versions
- Agent tracing & analysis
- Test Suites & assertions
- Agent Playground
Free Cloud
Free
- Up to 10 team members
- 25k spans per month
- 60-day data retention
- Agent tracing & analysis
- Test Suites & assertions
- Agent Playground
Pro Cloud
- $19 Per month
- Up to 50 team members
- 100k spans per month
- 60-day data retention
- Customizable monthly span limits
- Customizable data retention periods
Enterprise
Price on request
- Unlimited team members
- Custom usage plans
- Flexible deployments
- Service accounts and view-only users
- Single sign-on
- Dedicated support and SLAs