Both trace multi-step agents, score production traffic with LLM judges or code, queue runs for human review and version prompts. Opik puts the whole platform under Apache-2.0 and lets you run it yourself at no cost; LangSmith is a hosted, per-seat product with the closest fit to LangChain, LangGraph and Deep Agents. Licence and framework usually decide it.[1][4]
LangSmith is LangChain's proprietary platform, priced per seat, with self-hosting only on an Enterprise contract; Opik is Comet's Apache-2.0 platform that you can self-host for free or use as a cloud priced by spans. Both trace agents, run offline and online evals and version prompts. LangSmith fits teams on LangChain or LangGraph; Opik fits teams that want the whole platform as open source.
| Decision factor | LangSmith | Opik |
|---|---|---|
| Ownership | LangChain, independent; $125M Series B in October 2025 | Comet ML, independent; reports $70M in total funding |
| Licence | Proprietary platform; client SDKs are MIT | Apache-2.0, covering tracing, evaluation, UI, backend and SDKs |
| Self-hosting | Enterprise add-on with a licence key, on Kubernetes; BYOC also Enterprise | Free; Kubernetes for production, Docker for local use; user administration excluded |
| Run-time checks | LLM Gateway (public beta): spend caps, model access and redaction on model calls | Guardrails your code calls; a failed check raises an exception |
| Prompt tooling | Prompt and Context Hub with commit-based versions, tags and a playground | Prompt Library with environment-tagged versions, playground and Prompt Optimizer |
| OpenTelemetry | OTLP ingest; maps OpenInference, OpenLLMetry and Traceloop attributes | OTLP over HTTP only; gRPC documented as causing errors |
| Free tier | Developer: 1 seat, 5k base traces a month, 14-day retention | Free Cloud: 10 members, 25k spans a month, 60-day retention |
| First paid tier | Plus: $39 per seat a month, 10k base traces | Pro Cloud: $19 a month, 100k spans, then $5 per 100k |
| Capability | LangSmith | Opik | Prefactor |
|---|---|---|---|
| Observe | |||
| LLM call tracing | ✓ | ✓ | ✓ |
| Multi-step agent tracing (tool calls, nested spans) | ✓ | ✓ | ✓ |
| Cost tracking | ✓ | ✓ | ✓ |
| OpenTelemetry support | ✓ | ✓ | ✓ |
| Evaluate | |||
| Offline evals (datasets, experiments) | ✓ | ✓ | ✓ |
| Online evals on production traffic | ✓ | ✓ | ✓ |
| LLM-as-a-judge scoring | ✓ | ✓ | ✓ |
| Code-based evaluators | ✓ | ✓ | ✓ |
| Human review and annotation | ✓ | ✓ | ✓ |
| Prompt versioning and playground | ✓ | ✓ | — |
| Act on findings | |||
| Runtime controls on live agent actions | Partial | Partial | ✓ |
| Gateway or proxy in the request path | — | Partial | — |
| Licence and deployment | |||
| Open source | Partial | ✓ | Partial |
| Self-hosting | ✓ | ✓ | Partial |
The clearest split between these two is who can run the platform. Opik's core, including tracing, evaluation, the UI, the backend and the SDKs, is in the public comet-ml/opik repository under Apache-2.0. Self-hosted Opik includes every feature except user administration. Comet describes its local Docker install as not production-ready and points production use to Kubernetes.
LangSmith is proprietary. Its client SDKs are MIT-licensed, and on 16 October 2026 LangChain archived the old langsmith-sdk monorepo and split it into separate Python and JavaScript repositories. Running LangSmith yourself is an add-on to the Enterprise plan, with a licence key and a Kubernetes deployment through Helm or Terraform. A BYOC option, where LangChain runs the control plane and you run the data plane, is also Enterprise only.
If traces must stay inside your network and there is no budget for an Enterprise contract, Opik is the option that works without one.
LangSmith is built by the team behind LangChain, LangGraph and Deep Agents, and its integration with those frameworks is the deepest. It is not limited to them: its documentation lists the OpenAI Agents SDK, Claude Agent SDK, CrewAI, Google ADK, Mastra, Pydantic AI and others, plus decorators and wrappers for plain code. LangSmith also sells agent deployment (formerly LangGraph Platform), and Engine, in public beta since May 2026, reads production traces and drafts proposed fixes.
Opik lists more than 30 framework and provider integrations, among them LangGraph, CrewAI, OpenAI Agents, Google ADK, Claude Agent SDK, DSPy and n8n. Its SDKs cover Python, TypeScript, Java and .NET, with Ruby through OpenTelemetry. Its evaluation set includes more than 30 built-in metrics and an 'Agent as a Judge' that reads traces with tools, and its Prompt Optimizer offers six optimisation algorithms.
Both accept OpenTelemetry. Opik takes OTLP over HTTP only and does not state which semantic conventions it supports; LangSmith maps OpenInference, OpenLLMetry and Traceloop attributes.
Human review works in a similar way in both. LangSmith's annotation queues support single-run and pairwise review, rubrics and required reviewer counts. Opik's queues let subject-matter experts score traces and, since October 2026, fill automatically from feedback-score conditions.
Each offers a partial run-time check, and they sit in different places. LangSmith's LLM Gateway, in public beta since 30 July 2026, sits in the model-call path and can cap spend (returning HTTP 402), restrict which models are used and redact personal data or secrets. It does not cover tool calls; LangChain describes tool and MCP gateways as planned.
Opik's guardrails (PII, topic, prompt injection, LLM judge and custom classifier) run as a blocking validate() call that fails closed. Your application code has to call them and handle the exception. PII and topic checks are English only, and self-hosted guards need a separate guardrails server, with a GPU recommended. Opik's own LLM Gateway is in beta and documented for development and testing only. Neither product documents approval holds on an agent's actions.
As of 20 October 2026, LangSmith prices per seat plus trace volume. Developer is free for one seat and 5k base traces a month with 14-day base retention. Plus is $39 per seat a month with 10k base traces, and the per-trace overage rate is not published. Other products, such as Engine and Deployment, are metered in LangChain Standard Units.
Opik prices by spans. Self-hosting is free with unlimited spans. Free Cloud covers up to 10 team members and 25k spans a month; Pro Cloud is $19 a month for up to 50 members and 100k spans, then $5 per 100k, with 60-day retention by default. A large team pays more under per-seat pricing; a high-traffic agent pays more under span pricing. A trace and a span are different units, so model your own volumes against LangSmith's pricing and Opik's pricing.
We sell the layer this section describes. Read it with that in mind.
Stated plainly, since Prefactor publishes this page: Prefactor is an agent evaluation platform that scores each step of a run and acts on what it finds through runtime policies on tool calls and data access, approval holds for risky actions, and a kill switch for a run, an agent or a team. LangSmith's gateway covers model calls only, and Opik's guardrails depend on your code calling them; neither holds an action for approval. Prefactor has no prompt playground or prompt registry and its platform is hosted, so if prompt iteration or free self-hosting is what you need, LangSmith or Opik is the better pick. See the full comparison of 15 platforms.
Reviewed against public product and documentation pages on October 20, 2026. If a vendor has changed a feature, product name, or positioning since then, send a correction and we will update it. Numbered source links in the page body point to the ordered sources below.
Prefactor scores each step of every agent run and can hold or stop an action when a score says it should not happen. Start on the free tier: 5,000 spans a month, no credit card required.
Prefactor helps teams observe, evaluate, and improve their AI agents in production, across every framework and provider.