Skip to main content
usedby

LLM observability & evals: the tools companies use

Tools that trace, test, evaluate and guard LLM apps and AI agents in production.

Compare tools

153

companies on usedby use a tool of this category

153 companies on usedby use at least one of the 6 tools in this category with a public customer. LangSmith leads with 43 companies. Then come Braintrust (42) and Arize AI (35). 6 companies run two of these tools or more. The most common pair is LangSmith with Galileo (2 companies).Latest check Oct 14, 2026: 149 of 160 public lines checked. Counts are of companies on usedby, not of the whole market: we list a company when a public page shows its AI tools.

Ranked by customers

Companies with a public line for each tool. Job-post signals count and are drawn apart in the proof bar.

LLM observability & evals tools ranked by companies on usedby
#ToolCustomersProof
1LangSmith43
2Braintrust42
3Arize AI35
4Comet Opik17
5Galileo13
6Fiddler AI10

6 companies run two of these tools or more. The most common pair is LangSmith with Galileo, at 2 companies.

LangSmith

LangSmith is used by 43 companies on usedby, 0 confirmed by a case study, a named quote or a press release. Customers include Nvidia, Toyota, and Cisco. Who uses LangSmith, Arize AI vs LangSmith

Customer page: langchain.com. Latest check of 37 of 43 lines: Oct 13, 2026.

Braintrust

Braintrust is used by 42 companies on usedby, 12 confirmed by a case study, a named quote or a press release. Customers include Cloudflare, Coursera, and Dropbox. Who uses Braintrust, Braintrust vs Browserbase

Customer page: braintrust.dev. Latest check of 42 of 42 lines: Oct 14, 2026.

Arize AI

Arize AI is used by 35 companies on usedby, 7 confirmed by a case study, a named quote or a press release. Customers include Booking.com, Tripadvisor, and PagerDuty. Who uses Arize AI, Arize AI vs LangSmith

Customer page: arize.com. Latest check of 30 of 35 lines: Oct 5, 2026.

Comet Opik

Comet Opik is used by 17 companies on usedby, 10 confirmed by a case study, a named quote or a press release. Customers include Uber, NTT Data, and Ancestry. Who uses Comet Opik, Comet Opik vs LangSmith

Customer page: comet.com. Latest check of 17 of 17 lines: Oct 7, 2026.

Newest customers

Companies first listed by a tool of this category in the last 90 days.

No new customer yetTracking of changes on these tools' pages started on October 3, 2026. New customers appear here, dated, after two complete checks of those pages.

Most compared

Side-by-side pages with the companies behind each tool.

Questions about LLM observability & evals

Answers drawn from usedby data.

What is the most used tool in LLM observability & evals?

LangSmith, used by 43 of the 153 companies on usedby in this category. Then come Braintrust (42) and Arize AI (35). Who uses LangSmith.

Do more companies use LangSmith or Braintrust?

LangSmith: 43 companies on usedby, against 42 for Braintrust.

Do companies use several tools in LLM observability & evals?

Yes: 6 of the 153 companies on usedby in this category run two or more. The most common pair is LangSmith with Galileo (2 companies).

How does usedby rank these tools?

By the number of companies with a public line for each tool, read from dated captures of vendor customer pages, case studies, press releases and job posts. Every proof level counts in the totals and the ranking, job-post signals included (the weakest proof, shown apart in the proof bar); ties go to the stronger proofs. usedby lists a company when a public page shows it, so these counts describe the companies on usedby, not the whole market.