LangSmith vs Arize Phoenix in 2026 - Turnkey and Pricey vs Open and OTel-Native

LangSmith and Arize Phoenix sit at opposite corners of the observability market, which is exactly why people put them head to head. LangSmith is the turnkey, closed, deeply integrated option - if you build on LangChain, nothing traces it better. Phoenix is the open, OpenTelemetry-native option you can run on your laptop in under a minute. One is a premium managed platform coupled to a framework; the other is a source-available project you self-host. The right pick depends almost entirely on your stack and your budget. I have added Langfuse as the third reference, because it is the cheap self-hostable middle a lot of teams land on.

The short version

Tool	Best for	Self-host	License	Starting price
LangSmith	LangChain and LangGraph apps	Enterprise only	Closed	$39/seat/mo
Arize Phoenix	Fast OSS tracing and RAG eval	Full, free	ELv2 server, Apache client	Free (OSS)
Langfuse	Cheap self-host, framework-agnostic	Full, free	MIT	Free / $29/mo

LangSmith: the deepest LangChain tracing, at a price

LangSmith is LangChain’s own platform, and its coupling to the LangChain and LangGraph stack is the whole pitch. Add a callback and every chain, tool call and agent step shows up traced with zero extra instrumentation. No other platform gives you that depth on LangChain code, because no other platform is built by the people who ship LangChain. The eval side is strong too - LLM-as-judge, custom evaluators, dataset management, and Align Evals, a genuinely useful workflow for calibrating your LLM judge against human scores. It is backed by a $1.25B company (roughly $260M raised), so survival is not the question.

Two things keep it from being an easy default. The trace bill runs roughly $2,514/mo at 1M base traces on one seat - about 25x Langfuse - with base traces at $2.50 per 1,000 and extended at $5.00. And you cannot self-host your way out. LangSmith is fully closed source, and self-hosting is Enterprise-only behind a sales call. It accepts OpenTelemetry as a receiver, so you are not forced onto the LangChain SDK, but the zero-config magic only appears when you use it. If you are not on the LangChain stack, you are paying premium rates for a generic tracer.

Arize Phoenix: fastest to try, best RAG eval, read the license

Arize Phoenix is the OSS pick that starts fastest - a working trace UI on your laptop in under a minute. It is genuinely OpenTelemetry-native, built on OTel and Arize’s own OpenInference conventions, so it is framework-agnostic in a way LangSmith is not. Its 50+ pre-built eval metrics include what reviewers call the best RAG evaluation in the category - serious retrieval and answer scoring without writing your own judge prompts. Phoenix OSS is free with no usage caps and no feature gates on the actual features. Arize the company is well funded - a $70M Series C in February 2025 - so it is not going anywhere.

The gotcha is the license, and it is a real one. Arize markets Phoenix as “fully open source, no feature gates,” but the main server repo is Elastic License 2.0 - source-available, not OSI-approved open source. ELv2 forbids offering Phoenix as a hosted service to third parties. Only the client and eval subpackages - phoenix-evals, phoenix-client, phoenix-otel - are Apache-2.0. For internal use it behaves like open source; if your business model is reselling it as a service, the license rules that out. There are also reports of ingest lag before traces appear, so weigh that if real-time visibility matters. And note Arize AX, the separate managed cloud, is a different product with pricing you cannot read publicly - budget a sales call there.

Langfuse: the cheap, cleanly-licensed middle

If LangSmith is too expensive and closed, and Phoenix’s ELv2 server is a sticking point, Langfuse is the option that splits the difference. It is MIT-licensed - genuinely OSI open source - and self-hosts free with only three features enterprise-gated. It is framework-agnostic, runs as an OpenTelemetry backend, and at 1M events a month runs about $101/mo managed against LangSmith’s roughly $2,514. The catch is operational: the v3 self-host needs four services (Postgres plus ClickHouse, Redis and S3-compatible storage) and the migration is where people get stuck. Where Phoenix pulls ahead of Langfuse is eval depth, RAG especially, and the sub-minute local start. Where Langfuse pulls ahead is license clarity. Both are OTel-friendly.

LangSmith vs Arize Phoenix: which should you pick?

• You live in LangChain and LangGraph and the bill does not scare you - LangSmith. The zero-config tracing really is the deepest available for that stack, and Align Evals is a genuine reason to be here.
• You want fast, framework-agnostic OSS tracing you run yourself - Arize Phoenix, as long as you are not reselling it as a service.
• RAG evaluation is the priority - Phoenix has the best in the category, with 50+ metrics and no judge prompts to write.
• You are cost-sensitive at production volume - anything but LangSmith. Phoenix self-hosted is free; Langfuse is roughly 25x cheaper at scale and MIT-licensed.
• You need strict OSI-approved open source - Langfuse, not Phoenix. The ELv2 server is source-available, not OSI open source.

The honest summary: LangSmith wins on integration depth for one specific stack and loses everywhere else on cost and openness. Phoenix wins on speed-to-start, OTel-native architecture and RAG eval, with a license caveat that only bites resellers. If you build on LangChain and can absorb the bill, stay on LangSmith. Otherwise Phoenix - or Langfuse if you want a cleaner license - gives you the same job for a fraction of the money.

Every price and date here was read from each vendor’s own materials and verified on 26 July 2026. Arize AX pricing could not be confirmed - the vendor page returned an HTTP 403. This category ships breaking changes monthly, so we re-verify every 30 days.