--- title: Langfuse is the most used open-source LLM engineering platform seoTitle: "Langfuse adoption: the most used open-source LLM engineering platform" description: Live SDK download, GitHub star, and enterprise adoption metrics comparing Langfuse with other LLM observability and evaluation platforms. --- # Langfuse is the most used open-source LLM engineering platform Teams use [Langfuse](/) for [observability](/docs/observability/overview), [evaluation](/docs/evaluation/overview), and [prompt management](/docs/prompt-management/overview) of LLM applications and agents. This page collects the public metrics behind the claim. Most numbers below are live and refresh automatically. - **50,000+** Companies - **90B+** Observations / month - **65M+** SDK installs / month - **38M+** Docker pulls - **21** Of the Fortune 50 - **129** Of the Fortune 500 - **37,014** GitHub stars - **100+** Integrations Companies using Langfuse include [Ramp](/users/ramp), [Canva](/users/canva), Expedia Group, Pigment, [Khan Academy](/users/khan-academy), [Hugging Face](/users/hugging-face), Intuit, [SumUp](/users/sumup), [Merck](/users/merckgroup), Samsara, Cisco, and Rocket Money. ## SDK downloads [#sdk-downloads] Langfuse has the most downloaded SDKs among LLM observability and evaluation platforms on both PyPI and npm. | | PyPI downloads / month | npm downloads / month | | ---------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | 🪢 Langfuse | [![Langfuse PyPI downloads](https://img.shields.io/pypi/dm/langfuse?style=for-the-badge&label=%20&labelColor=black&color=orange)](https://pypi.org/project/langfuse/) | [![Langfuse npm downloads](https://img.shields.io/npm/dm/%40langfuse%2Ftracing?style=for-the-badge&label=%20&labelColor=black&color=orange)](https://www.npmjs.com/package/@langfuse/tracing) | | MLflow | [![MLflow PyPI downloads](https://img.shields.io/pypi/dm/mlflow?style=for-the-badge&label=%20&labelColor=black&color=grey)](https://pypi.org/project/mlflow/) | [![MLflow npm downloads](https://img.shields.io/npm/dm/mlflow-tracing?style=for-the-badge&label=%20&labelColor=black&color=grey)](https://www.npmjs.com/package/mlflow-tracing) | | Pydantic Logfire | [![Logfire PyPI downloads](https://img.shields.io/pypi/dm/logfire?style=for-the-badge&label=%20&labelColor=black&color=grey)](https://pypi.org/project/logfire/) | [![Logfire npm downloads](https://img.shields.io/npm/dm/logfire?style=for-the-badge&label=%20&labelColor=black&color=grey)](https://www.npmjs.com/package/logfire) | | Braintrust | [![Braintrust PyPI downloads](https://img.shields.io/pypi/dm/braintrust?style=for-the-badge&label=%20&labelColor=black&color=grey)](https://pypi.org/project/braintrust/) | [![Braintrust npm downloads](https://img.shields.io/npm/dm/braintrust?style=for-the-badge&label=%20&labelColor=black&color=grey)](https://www.npmjs.com/package/braintrust) | | Opik (Comet) | [![Opik PyPI downloads](https://img.shields.io/pypi/dm/opik?style=for-the-badge&label=%20&labelColor=black&color=grey)](https://pypi.org/project/opik/) | [![Opik npm downloads](https://img.shields.io/npm/dm/opik?style=for-the-badge&label=%20&labelColor=black&color=grey)](https://www.npmjs.com/package/opik) | | Arize Phoenix | [![Phoenix PyPI downloads](https://img.shields.io/pypi/dm/arize-phoenix-otel?style=for-the-badge&label=%20&labelColor=black&color=grey)](https://pypi.org/project/arize-phoenix-otel/) | [![Phoenix npm downloads](https://img.shields.io/npm/dm/%40arizeai%2Fphoenix-otel?style=for-the-badge&label=%20&labelColor=black&color=grey)](https://www.npmjs.com/package/@arizeai/phoenix-otel) | _Live via [shields.io](https://shields.io). Packages: `langfuse` and `@langfuse/tracing`, `mlflow` and `mlflow-tracing`, `logfire`, `braintrust`, `opik`, `arize-phoenix-otel` and `@arizeai/phoenix-otel`._ ## GitHub stars [#github-stars] Langfuse is the most starred open-source LLM engineering platform on GitHub. ![GitHub star history of Langfuse compared with other LLM engineering platforms](/images/adoption/star-history.svg) _Source: [star-history.com](https://star-history.com/#langfuse/langfuse&comet-ml/opik&arize-ai/phoenix&pydantic/logfire&Date)_ ## Customers [#customers] Teams at [Canva](/users/canva), [Khan Academy](/users/khan-academy), [Merck](/users/merckgroup), and [many more](/users) run Langfuse in production. > "We wanted something built for agents as users first. And that means API first. An agent should never get stuck waiting for a human because of a deficiency in the API. This is what Langfuse is." > > [Ramp](/users/ramp) > "Langfuse hits the sweet spot between engineering requirements and empowerment of non-technical users to contribute their domain expertise." > > [Canva](/users/canva) > "Langfuse has enabled our developers to get extremely fast feedback. It's fundamental to how our developers understand their AI implementations." > > [Khan Academy](/users/khan-academy) > "The whole pipeline to improve our chat agent happens entirely in Langfuse. Engineers and domain experts work hand-in-hand." > > [DKB](/users/dkb) > "Langfuse enables us to track every prompt, response, cost, and latency in real time, turning black-box models into auditable, optimizable assets." > > [Merck](/users/merckgroup) > "Building on Langfuse we saved 30% of external BPO cost by deflecting 50% of support conversations to AI." > > [SumUp](/users/sumup) > "You can connect Claude Code to your Langfuse MCP - or the CLI directly - and it starts fixing and trying until it is fixed." > > [Trade Republic](/users/trade-republic) > "In just a few minutes you're up and running, and you're already able to see what's going on." > > [Hugging Face](/users/hugging-face) > "Langfuse has made it a lot easier for us at Cresta to iterate on LLM behavior. It has sped up how quickly we can test, learn, improve the model, and build great AI agents." > > [Cresta](/users/cresta) > "Building a product like ours would be impossible without Langfuse." > > [Ravenna](/users/ravenna) > "Having this data set, having this whole test suite that a developer can run on their machine and quickly get results, it's a game changer for us." > > [Slite](/users/slite) > "I always say it's like the CMS of prompts. It's a game changer." > > [Evolve](/users/evolve) > "When a customer pings us, we instantly open Langfuse. This allows keeping support resolution under 8 minutes." > > [Magic Patterns](/users/magic-patterns-ai-design-tools) ## How Langfuse compares [#compare] Head-to-head comparisons on features, self-hosting, and pricing. - [Langfuse vs. LangSmith](/compare/langsmith) - [Langfuse vs. Braintrust](/compare/braintrust) - [Langfuse vs. Arize AX / Phoenix](/compare/arize-phoenix) - [Langfuse vs. MLflow](/compare/mlflow) - [Langfuse vs. Pydantic Logfire](/compare/logfire) - [All comparisons](/compare) ## Open-source projects using Langfuse [#open-source-projects] The 25 most-starred open-source projects that depend on the Langfuse Python or JS SDK. See the [full list](/faq/all/packages-depending-on-langfuse). | Repository | Stars | | :-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -----: | | ![avatar](https://avatars.githubusercontent.com/u/130738209?s=40&v=4)   [Significant-Gravitas](https://github.com/Significant-Gravitas) / [AutoGPT](https://github.com/Significant-Gravitas/AutoGPT) | 187903 | | ![avatar](https://avatars.githubusercontent.com/u/127165244?s=40&v=4)   [langgenius](https://github.com/langgenius) / [dify](https://github.com/langgenius/dify) | 160407 | | ![avatar](https://avatars.githubusercontent.com/u/85702467?s=40&v=4)   [langflow-ai](https://github.com/langflow-ai) / [langflow](https://github.com/langflow-ai/langflow) | 156333 | | ![avatar](https://avatars.githubusercontent.com/u/4158466?s=40&v=4)   [bytedance](https://github.com/bytedance) / [deer-flow](https://github.com/bytedance/deer-flow) | 84985 | | ![avatar](https://avatars.githubusercontent.com/u/320819803?s=40&v=4)   [cheaper-inference](https://github.com/cheaper-inference) / [OmniRoute](https://github.com/cheaper-inference/OmniRoute) | 84532 | | ![avatar](https://avatars.githubusercontent.com/u/131470832?s=40&v=4)   [lobehub](https://github.com/lobehub) / [lobehub](https://github.com/lobehub/lobehub) | 83873 | | ![avatar](https://avatars.githubusercontent.com/u/23818?s=40&v=4)   [abi](https://github.com/abi) / [screenshot-to-code](https://github.com/abi/screenshot-to-code) | 81022 | | ![avatar](https://avatars.githubusercontent.com/u/6154722?s=40&v=4)   [microsoft](https://github.com/microsoft) / [ai-agents-for-beginners](https://github.com/microsoft/ai-agents-for-beginners) | 78925 | | ![avatar](https://avatars.githubusercontent.com/u/14957082?s=40&v=4)   [openai](https://github.com/openai) / [openai-cookbook](https://github.com/openai/openai-cookbook) | 76955 | | ![avatar](https://avatars.githubusercontent.com/u/184127137?s=40&v=4)   [cline](https://github.com/cline) / [cline](https://github.com/cline/cline) | 71312 | | ![avatar](https://avatars.githubusercontent.com/u/121462774?s=40&v=4)   [BerriAI](https://github.com/BerriAI) / [litellm](https://github.com/BerriAI/litellm) | 62709 | | ![avatar](https://avatars.githubusercontent.com/u/130722866?s=40&v=4)   [run-llama](https://github.com/run-llama) / [llama_index](https://github.com/run-llama/llama_index) | 52806 | | ![avatar](https://avatars.githubusercontent.com/u/169401942?s=40&v=4)   [LibreChat-AI](https://github.com/LibreChat-AI) / [LibreChat](https://github.com/LibreChat-AI/LibreChat) | 46745 | | ![avatar](https://avatars.githubusercontent.com/u/67365215?s=40&v=4)   [volcengine](https://github.com/volcengine) / [OpenViking](https://github.com/volcengine/OpenViking) | 41535 | | ![avatar](https://avatars.githubusercontent.com/u/118165258?s=40&v=4)   [HKUDS](https://github.com/HKUDS) / [LightRAG](https://github.com/HKUDS/LightRAG) | 40529 | | ![avatar](https://avatars.githubusercontent.com/u/15967950?s=40&v=4)   [directus](https://github.com/directus) / [directus](https://github.com/directus/directus) | 38820 | | ![avatar](https://avatars.githubusercontent.com/u/34411969?s=40&v=4)   [DayuanJiang](https://github.com/DayuanJiang) / [next-ai-draw-io](https://github.com/DayuanJiang/next-ai-draw-io) | 36683 | | ![avatar](https://avatars.githubusercontent.com/u/131946000?s=40&v=4)   [onyx-dot-app](https://github.com/onyx-dot-app) / [onyx](https://github.com/onyx-dot-app/onyx) | 33212 | | ![avatar](https://avatars.githubusercontent.com/u/102226726?s=40&v=4)   [labring](https://github.com/labring) / [FastGPT](https://github.com/labring/FastGPT) | 29994 | | ![avatar](https://avatars.githubusercontent.com/u/149120496?s=40&v=4)   [mastra-ai](https://github.com/mastra-ai) / [mastra](https://github.com/mastra-ai/mastra) | 29335 | | ![avatar](https://avatars.githubusercontent.com/u/14985020?s=40&v=4)   [vercel](https://github.com/vercel) / [ai](https://github.com/vercel/ai) | 27740 | | ![avatar](https://avatars.githubusercontent.com/u/157326433?s=40&v=4)   [onlook-dev](https://github.com/onlook-dev) / [onlook](https://github.com/onlook-dev/onlook) | 27117 | | ![avatar](https://avatars.githubusercontent.com/u/137907881?s=40&v=4)   [promptfoo](https://github.com/promptfoo) / [promptfoo](https://github.com/promptfoo/promptfoo) | 26836 | | ![avatar](https://avatars.githubusercontent.com/u/170767358?s=40&v=4)   [kortix-ai](https://github.com/kortix-ai) / [CompanyOS](https://github.com/kortix-ai/CompanyOS) | 20378 | | ![avatar](https://avatars.githubusercontent.com/u/255961667?s=40&v=4)   [gastownhall](https://github.com/gastownhall) / [gastown](https://github.com/gastownhall/gastown) | 18691 | **Something out of date?** [Open a pull request](https://github.com/langfuse/langfuse-docs/blob/main/content/marketing/adoption.mdx) or [reach out](/support). --- ## Agent Instructions This page is part of the [Langfuse](https://langfuse.com) documentation, published as plain Markdown for AI agents. Every page is available as Markdown by appending `.md` to its URL, or by sending an `Accept: text/markdown` header. This page: `https://langfuse.com/adoption.md`. ### Querying these docs If the answer is not on this page, query the documentation instead of guessing: - **Semantic search** across all Langfuse docs, returning an answer with the relevant pages and excerpts. Ask a specific, self-contained question: ```bash curl -sG "https://langfuse.com/api/search-docs" --data-urlencode "query=How do I trace a LangGraph agent?" ``` - **Index of every page**: , with per-section indexes [llms-docs.txt](https://langfuse.com/llms-docs.txt), [llms-integrations.txt](https://langfuse.com/llms-integrations.txt), and [llms-self-hosting.txt](https://langfuse.com/llms-self-hosting.txt). ### Before writing Langfuse code - **Install the [Langfuse Agent Skill](https://langfuse.com/docs/api-and-data-platform/features/agent-skill).** It encodes Langfuse's own best practices for instrumentation, prompt management, and evaluation, and materially improves results. - **Read [What does a good trace look like?](https://langfuse.com/docs/observability/best-practices.md)** before instrumenting an application. - **Verify endpoints, parameters, and response fields** against the [API reference](https://api.reference.langfuse.com) instead of inferring them from code examples. - **Use the [Langfuse CLI](https://langfuse.com/docs/api-and-data-platform/features/cli)** (`npx @langfuse/cli api `) to read or write traces, prompts, datasets, and scores from the terminal. Found an error in these docs? Please open an issue at .