Skip to the content
Know your tech before you buy it. Makers: get listed from $49/mo

Opik

comet.com

  • Free plan
  • Open source
  • Windows
  • macOS
  • Linux
  • Web
  • API
  • Self-hosted
8.7 Club score Visit comet.com

What Opik is

Opik helps teams building applications or agents that call an LLM log agent actions and evaluate behavior across development, testing, and production. It captures and visualizes traces, supports annotation, and provides audit logs. Its evaluation tools include more than 30 metrics covering outcomes such as answer relevance, context precision, task completion, and hallucination. In production, Opik can evaluate traces in real time, alert when criteria fail, and track token usage and model costs. Test Suites provide pass/fail assertions, an Agent Playground supports end-to-end tests, and a Prompt Optimizer offers six algorithms. The core observability and evaluation features are available as open-source code for local use. Opik includes a REST API and Python client, with Python and TypeScript SDKs listed, plus integrations with more than 40 AI frameworks, model providers, and gateways. Free Cloud allows up to 10 team members, 25,000 spans per month, and 60-day retention. Pro Cloud costs $19 per month and includes up to 50 team members and 100,000 spans monthly.

Who it suits

Opik suits teams building applications or agents that call an LLM and need tracing, evaluation, or production monitoring. Its free options can serve development and testing, while local self-hosting is available through open-source code.

Good to know, and what to watch

  • Includes more than 30 evaluation metrics.
  • Evaluates production traces in real time.
  • Tracks token usage and model costs.
  • Open-source core can be run locally.
  • Integrates with more than 40 frameworks, providers, and gateways.
  • Free Cloud is limited to 25,000 spans monthly.
  • Free Cloud retains data for 60 days.
  • Pro Cloud costs $19 per month.

Enterprise

—

Free Cloud

Free

Open Source

Free

Pro Cloud

$19/mo

Free plan

Getting started

  1. Open the Opik product website
  2. Choose Free Cloud, Pro Cloud or Enterprise
  3. Download and run the open-source edition locally if self-hosting
  4. Connect through the REST API, Python client or listed SDKs
  5. Add an integration such as LangChain, OpenAI, Google ADK or LangGraph
  6. Create traces, evaluations or test suites

Limits to check

Free Cloud allows up to 10 team members, 25,000 spans per month and 60-day data retention. Pro Cloud raises the limits to 50 members and 100,000 spans monthly, with customizable span limits and retention periods.

The Club's review

Overview

Opik is an observability and evaluation platform for applications and agents that call an LLM. It is most compelling for teams that want to connect agent testing with production oversight, while retaining the option to run the open-source version locally.

Its broad workflow—from trace review to evaluation and cost tracking—makes it a practical choice for teams building LLM applications. Cloud users should weigh the monthly span limits and retention terms against their expected traffic.

Key features

Opik captures and visualizes agent actions, with trace annotation and audit logs for investigation. More than 30 evaluation metrics address outcomes such as answer relevance, context precision, task completion, and hallucination. That range gives teams multiple ways to assess behavior, though they still need to select criteria that reflect their own application.

Production monitoring can evaluate traces in real time, alert when criteria fail, and track token usage and model costs. Test Suites provide pass/fail assertions, the Agent Playground supports end-to-end tests, and the Prompt Optimizer offers six algorithms. The combination suits teams that want evaluation to continue beyond development rather than treating it as a pre-release exercise.

The core observability and evaluation feature set is open source and can be run locally. Opik includes a REST API and Python client; its pricing page also lists Python and TypeScript SDKs and more than 40 integrations across AI frameworks, model providers, and gateways. Named examples include LangChain, OpenAI, Google ADK, and LangGraph.

Pricing

Free Cloud costs nothing (billed $0) and allows up to 10 team members, 25,000 spans per month, and 60-day data retention. It is a useful starting point for development and testing, but the span cap may constrain teams with sustained production traffic.

Open Source costs nothing (billed $0) and provides the full AI observability and agent testing feature set on the same codebase as the hosted versions, for self-hosting. It avoids a hosted-plan price, but requires teams to run it themselves.

Pro Cloud costs $19 a month (billed $19 per month). It raises the limits to 50 team members and 100,000 spans per month, retains data for 60 days, and allows span limits and retention periods to be customized. Email support is included, alongside community and self-service support. This is the clearest fit for teams that need more cloud capacity than Free provides; the listed member and span ceilings still matter.

Enterprise has custom pricing, unlimited team members, custom usage plans, flexible deployments, dedicated support, and SLAs. It is aimed at larger or more regulated organizations: enterprise SSO and SOC 2, ISO 27001, ISO 9001, HIPAA, and GDPR compliance are listed for this plan. Comet's Trust Center lists ISO 9001:2015, ISO/IEC 27001:2022, and SOC 2 Type 2 compliance.

A free trial is available. Community and self-service support are offered across plans; Pro and Enterprise add email support, while Enterprise includes dedicated support and SLAs.

Platforms

Opik supports API, Linux, self-hosted, and web deployments. That mix gives teams a hosted route as well as a local deployment option, rather than tying adoption to cloud alone.

Who it's for

Opik is built for teams developing applications or agents that call an LLM. The Free Cloud plan suits development and testing with modest span volume; Pro is a step up for teams needing more cloud capacity. Self-hosting is the better fit for teams that want the full feature set under their own deployment, while Enterprise is positioned for organizations needing flexible usage, deployment, and support terms.

Pros and cons

  • Pros: Tracing, evaluation, testing, and production monitoring sit in one product, helping teams follow agent behavior across stages.
  • Pros: The full core feature set is open source and self-hostable, with more than 40 integrations and named support for frameworks and providers such as LangChain and OpenAI.
  • Pros: Free Cloud includes up to 10 team members and 25,000 monthly spans, giving small teams room to develop and test without a bill.
  • Cons: Free Cloud's 25,000-span monthly ceiling and 60-day retention can be restrictive for heavier production workloads.
  • Cons: Pro Cloud's 100,000 monthly spans and 50-member cap may still require a custom usage plan as a team or workload grows.
  • Cons: The self-hosted option means teams must operate their own deployment rather than relying on the hosted service.

Alternatives

W&B Weave is another freemium option, with its Free plan including 1 GB/mo Weave data ingestion, 5 GB/mo storage, and AI application evaluations, tracing, and scorers; consider it when those stated ingestion and storage allowances suit your needs.

Helicone is a freemium, open-source alternative for teams considering another self-hosted or API-based option.

Maxim AI offers a free Developer plan for up to 3 seats, 1 workspace, 10k logs per month, and 3-day data retention; it may suit a smaller team willing to work within those limits.

Respan (formerly Keywords AI) has a free plan with 100k logs, 1k scores, 5 datasets, 2 evaluators, and 5 prompts; consider it if those included quantities match your evaluation workflow.

OpenLIT offers an OSS plan with unlimited self-hosted usage, users, projects, and environments, making it a fit for teams prioritizing those uncapped self-hosted allowances.

Agenta has a free Hobby plan for 2 team members, 5,000 agent runs per month, 20 evaluations per month, and one-week trace retention; it may suit a small team whose usage fits those caps.

Portkey offers a free Developer plan with 10k recorded logs per month, 3-day log retention, 30-day metrics retention, and 3 prompt templates; it may fit teams whose needs align with those limits.

Fiddler AI is another freemium option for teams comparing AI observability and guardrails.

For more options, browse AI LLM Observability Tools, AI LLM Evaluation Tools, Model Monitoring Software, MLOps Platforms, AI Agent Evaluation Tools, and LLM Observability Tools.

Verdict

Choose Opik if your team needs a joined-up workflow for agent traces, evaluation, testing, and production monitoring, especially if self-hosting matters. Its open-source core and generous development-oriented Free Cloud allowance make it easy to start; teams with growing traffic should look elsewhere or plan for a higher-capacity arrangement if the cloud span caps and retention do not fit.

The verdict

Opik combines agent tracing, evaluation, testing tools, and production monitoring, with cloud and open-source deployment options. Check span limits and retention against your workload before choosing a plan.

Questions people ask

Is Opik free?

Yes. Free Cloud costs nothing, and the open-source edition is also free. Pro Cloud costs $19 a month.

Can Opik be self-hosted?

Yes. The open-source feature set can be downloaded and run locally, and self-hosted deployment is listed.

Which platforms and interfaces are available?

Opik supports API, Linux, self-hosted and web deployments. It includes a REST API, Python client, and Python and TypeScript SDKs.

What are the Free Cloud quotas?

Free Cloud supports up to 10 team members, 25,000 spans per month and 60-day data retention.

Does Opik track LLM costs?

Yes. It tracks token usage and model costs, including token cost tracking in production monitoring.

What support comes with the plans?

Community and self-service support are listed across plans. Pro and Enterprise add email support, while Enterprise includes dedicated support and SLAs.

What the maker publishes

Agent optimization
Opik includes prompt and tool optimization with multiple optimization algorithms and an optimization dashboard. source
Annotation
Reviewers can annotate traces or complete conversations using custom feedback schemas and annotation queues. source
API and SDKs
The product includes a REST API and Python client, and the pricing page lists Python and TypeScript SDKs. source
API and SDKs
Opik offers a REST API, Python and TypeScript SDKs, and OpenTelemetry ingestion for languages including Java, Ruby and .NET. source
Audience
The pricing page describes Opik as a platform for teams building applications or agents that make calls to an LLM, and says its free plans suit development and testing. source
Company profile
Comet is a privately held software-development company headquartered in New York with 51–200 employees and a 2017 founding year. source
Company usage claim
Comet says more than 150,000 builders use Opik. source
Documentation
Documentation includes feature guides, integration pages, self-hosting instructions, SDK references, REST API references, cookbooks and a changelog. source
Enterprise security
The pricing page lists enterprise SSO and SOC 2, ISO 27001, ISO 9001, HIPAA, and GDPR compliance for the Enterprise plan. source
Evaluation
The product page says Opik includes more than 30 metrics for evaluating answer relevance, context precision, task completion, hallucination, and other outcomes. source
Evaluation
Opik supports datasets, experiments, assertions and LLM-as-a-judge metrics for issues such as hallucination, relevance and moderation. source
Founded
2017 source
Framework integrations
Named integrations include LangChain, LangGraph, LlamaIndex, CrewAI, AutoGen, Google ADK, OpenAI Agents, Semantic Kernel and many others. source
Free plan limits
The Free Cloud plan lists up to 10 team members, 25,000 spans per month, and 60-day data retention. source
Funding
TechCrunch reported a $50 million Series B in November 2021, bringing Comet's reported total raised to almost $70 million. source
Guardrails
Self-hosted deployments can use PII, topic and custom guardrails to detect or block unsafe inputs and outputs. source
Headquarters
New York City, United States source
Integrations
The pricing page describes integrations with more than 40 AI frameworks, model providers, and AI gateways, naming LangChain, OpenAI, Google ADK, and LangGraph as examples. source
Integrations
The integration catalog covers TypeScript, Python, Java, .NET and Ruby ecosystems plus no-code tools and OpenTelemetry. source
Launch
Comet announced Opik as an open-source LLM evaluation platform in September 2024. source
License
The repository and full platform are released under the Apache-2.0 license. source
Limit: interface
The Slashdot reviewer said the user interface could be made more attractive. source
Limit: learning curve
The Gartner reviewer said it took time to fully understand the product. source
MCP extensibility
Its MCP server lets Claude Code, Cursor, VS Code Copilot, Codex and opencode query traces, score outputs and run evaluations from chat. source

Checked against the maker's pages on 21 Oct 2026.