14 open-source alternatives ranked by real activity

Open-source Weights & Biases alternatives

A curated, ranked list of the 14 best open-source alternatives to Weights & Biases.

The best open-source alternative to Weights & Biases is Langfuse. If that doesn't suit you, other good options are TraceRoot, Opik, Arize Phoenix and MLflow.

Commercial alternatives to Weights & Biases include Dataiku, DataRobot, Amazon SageMaker and Domino Data Lab. See all 25.

Weights & Biases alternatives are mainly AI infrastructure tools, but some are also AI tools and workflow automation tools. 9 of them shipped code in the last 30 days, 14 can be self-hosted, and 11 use a permissive licence.

Last updated November 13, 2026 · ranked by GitHub stars, growth and recent commits

Langfuse

An open-source LLM engineering platform for tracing, evaluating and improving AI applications, with prompt management, datasets and a playground, self-hosted or cloud.

GitHub stars
37k
Last commit
today
Latest release
v4.79.0
Self-hosted
Yes
Hosted version
Available
langfuse.comLangfuse homepage screenshot

Langfuse is an open-source LLM engineering platform that helps teams develop, monitor, evaluate and debug AI applications together. You instrument your app, and Langfuse records traces of LLM calls and surrounding logic such as retrieval, embeddings and agent actions, so complex runs and user sessions can be inspected and debugged.

Prompt management lets you centrally version and iterate on prompts, with caching on server and client so that changes do not add latency to your app. Evaluation features cover LLM-as-a-judge, code evaluators, user feedback, manual labeling and custom pipelines through the API and SDKs. Datasets provide test sets and benchmarks, there is an LLM playground for trying prompts, and integrations exist for OpenAI, LangChain and LlamaIndex. It is built on the ClickHouse database, and the Langfuse team has been part of ClickHouse since January 2026.

You can use Langfuse Cloud or self-host it, which the maintainers say takes minutes. The repository license is listed as 'Other' because it combines open-source and enterprise components, so check the terms. It suits teams shipping production LLM features who need visibility into quality, latency and cost.

Key features

  • Tracing of LLM calls, retrieval and agent actions
  • Prompt versioning with caching
  • LLM-as-a-judge and custom evaluations
  • Datasets for tests and benchmarks
  • Interactive LLM playground
  • Integrations with OpenAI, LangChain and LlamaIndex

Pricing: Core has a $29 monthly base that includes 100k units, then $8 per 100k units with volume discounts. Hobby and Pro tiers exist but their prices are not shown in the text; Enterprise is by sales.

Read more about LangfuseWebsiteGitHub#1 Weights & Biases alternative

TraceRoot

TraceRoot is an open-source observability layer for AI agents that turns production traces into detector findings, datasets and evaluations.

GitHub stars
1.4k
Last commit
yesterday
Latest release
v1.3.8
Self-hosted
Yes
Hosted version
Available
traceroot.aiTraceRoot homepage screenshot

TraceRoot describes itself as an open-source self-improving layer for AI agents. It captures traces from production agent systems and turns them into actionable feedback and evaluations, closing a loop with your coding agent so that problems found in production lead to improvements. The project is a Y Combinator S25 company.

Its capabilities include tracing of LLM calls, tool use and agent steps with OpenTelemetry-compatible Python and TypeScript SDKs, covering inputs, outputs, latency, tokens and cost. Detectors screen incoming traces for behaviors such as hallucinations, tool failures and logic errors, with configurable sampling and judge models. Datasets and evaluations let you version test cases, score runs and compare candidate versions, while dashboards and threshold alerts track quality, latency and cost.

A CLI exports traces and findings into your coding workflow, and an in-app AI assistant can explore traces with access to your source code and GitHub context, using a hosted model or your own key. TraceRoot offers a hosted cloud and a self-hosting guide, is written in TypeScript, and its license is listed as Other.

Key features

  • OpenTelemetry-compatible agent tracing
  • Detectors that flag failures in traces
  • Versioned datasets and evaluations
  • Dashboards and threshold alerts
  • CLI for exporting traces and findings
  • In-app AI assistant with code context
Read more about TraceRootWebsiteGitHub#2 Weights & Biases alternative

Opik

Opik is an open-source platform from Comet for tracing, evaluating and monitoring LLM applications, RAG systems and AI agents.

GitHub stars
23k
Last commit
yesterday
Latest release
2.2.143
Licence
Apache-2.0
Self-hosted
Yes
Hosted version
Available
comet.comOpik homepage screenshot

Opik is an open-source LLM observability and evaluation platform built by Comet. It covers the application lifecycle starting with early traces during development and ending with production monitoring, for teams that build LLM apps and AI agents. The code is Python, licensed under Apache-2.0, and the README says the full platform is free to self-host. The project documentation lives on the Comet site.

Capabilities include deep tracing of LLM calls and agent activity including complete trace trees for agents with several steps and tool calls, and evaluation with datasets, experiments and LLM-as-a-judge metrics for tasks like hallucination detection, moderation and RAG assessment. An Agent Optimizer SDK improves prompts and agents, dashboards and online evaluation rules support production monitoring, guardrails help with safe AI practices, and a PyTest integration tests LLM pipelines on each commit. Integrations include frameworks such as LangChain, LlamaIndex and OpenAI clients.

Key features

  • Tracing for LLM calls and agent steps
  • Datasets and experiments for evaluation
  • Evaluation metrics using LLM-as-a-judge
  • Prompt and agent optimization SDK
  • Production dashboards and online evaluation
  • PyTest integration for CI checks

Pricing: Free to self-host, with a free hosted tier of 25k spans. Pro includes 100k spans with extra spans at $5 per 100k; Enterprise is custom.

Read more about OpikWebsiteGitHub#3 Weights & Biases alternative

Arize Phoenix

Arize Phoenix is an open-source AI observability platform for tracing, evaluating and experimenting with LLM applications.

GitHub stars
12k
Last commit
yesterday
Latest release
@arizeai/phoenix-cli@1.22.0
Self-hosted
Yes
Hosted version
Available

Phoenix is built for experimentation, evaluation and troubleshooting of LLM apps. It traces an application's runtime using OpenTelemetry-based instrumentation, runs LLM-based evaluations of responses and retrieval, and keeps versioned datasets and experiments for comparing changes to prompts, models and retrieval.

Beyond tracing, it has a playground for comparing models and replaying traced calls, prompt management with version control and tagging, PXI, an agent built in for debugging traces, and a remote MCP server so tools such as Claude Code and Cursor can query traces, datasets and experiments. It is vendor and language agnostic, with support for frameworks like LangGraph, CrewAI, LlamaIndex and DSPy and providers such as OpenAI, Anthropic and AWS Bedrock.

Phoenix is open source and self-hosted. Install with pip install arize-phoenix and phoenix serve, run it with uvx, or deploy container images on Docker or Kubernetes through the Helm chart. For a managed platform, Arize sells Arize AX, and the same OpenTelemetry and OpenInference instrumentation works with both.

Key features

  • OpenTelemetry-based tracing of LLM applications
  • LLM-powered evaluation of responses and retrieval
  • Versioned datasets for experiments and fine-tuning
  • Prompt playground with model comparison and trace replay
  • Prompt management with version control and tagging
  • Remote MCP server for coding agents
  • Setup command detects framework and wires up trace export
Read more about Arize PhoenixWebsiteGitHub#4 Weights & Biases alternative

MLflow

MLflow is an open-source platform for debugging, evaluating and monitoring AI agents, LLM applications and machine learning models.

GitHub stars
29k
Last commit
yesterday
Latest release
v3.18.0
Licence
Apache-2.0
Self-hosted
Yes
Hosted version
Available
mlflow.orgMLflow homepage screenshot

MLflow is an open-source AI engineering platform for agents, LLM applications and ML models. It helps teams debug, evaluate, monitor and optimize AI applications in production while keeping control of costs and access to models and data. It is written in Python, released under the Apache-2.0 license, and has been developed since 2018, with topics covering MLOps, model management and LLMOps.

For LLM and agent work it offers production observability with traces, plus evaluation, prompt management and prompt optimization, and an AI Gateway that governs costs and model access. It works with Python, TypeScript and JavaScript, Java and other languages, and integrates with OpenTelemetry and MCP. Getting started means starting an MLflow server, enabling logging in your code and running it, then exploring traces and metrics in the web UI on port 5000. A setup wizard can connect to an MLflow server or a Databricks workspace and add tracing with a coding agent.

Key features

  • Tracing and observability for LLM apps and agents
  • Evaluation tools for models and agents
  • Prompt management and prompt optimization
  • AI Gateway for cost and model access control
  • OpenTelemetry and MCP integration
  • Web UI for exploring traces and metrics

Pricing: Free and open source under the Apache-2.0 license; the README also mentions connecting to a Databricks workspace.

Read more about MLflowWebsiteGitHub#5 Weights & Biases alternative

Dvc

DVC is an open-source version control system for data science and machine learning projects that applies software engineering best practices to data management.

GitHub stars
16k
Last commit
4 days ago
Latest release
3.67.1
Licence
Apache-2.0
dvc.orgDvc homepage screenshot

DVC is a free and open-source version control system designed for data science and machine learning projects. It uses a Git-like model to manage data, models, and experiments, bringing software engineering best practices to data teams.

The tool is intended for individual data scientists working on small projects, offering a Git extension that integrates with existing workflows. It also provides a VS Code extension for easier use.

Key features

  • Data version control using a Git-like model
  • Management of data, models, and experiments
  • Integration with Git for local workflows
  • VS Code extension available
Read more about DvcWebsiteGitHub#6 Weights & Biases alternative

Laminar

Open-source observability platform for AI agents with OpenTelemetry-based tracing, evaluations, dashboards, and plain-English alerts on agent behavior.

GitHub stars
3.5k
Last commit
yesterday
Latest release
v0.3.4
Licence
Apache-2.0
Self-hosted
Yes
Hosted version
Available
laminar.shLaminar homepage screenshot

Laminar is an open-source observability platform built for AI agents. It records what agents do during each run so that teams can debug failures, track behavior, and measure quality. The project is released under the Apache-2.0 license and the backend is written largely in Rust, with a TypeScript front end.

Tracing is based on OpenTelemetry, and a single line of SDK setup can automatically trace libraries such as the Vercel AI SDK, Browser Use, Stagehand, LangChain, OpenAI, Anthropic, and Gemini. Signals let you describe a behavior, such as an agent getting stuck in a loop, in plain English and be notified in Slack when it occurs. Evals run through an SDK and CLI locally or in CI, with a UI for comparing results, and SQL queries and dashboards cover traces, metrics, and events.

Coding agents can reach the data through MCP and CLI access to investigate issues themselves. Laminar can be self-hosted locally with Docker Compose, or used as the managed platform at laminar.sh. It suits teams building production agents who need tracing, evaluation datasets, and annotation in one place.

Key features

  • OpenTelemetry-native tracing SDK
  • Plain-English signals with Slack alerts
  • Evals via SDK and CLI with comparison UI
  • SQL queries over traces, spans, and metrics
  • Custom dashboards for traces and events
  • MCP and CLI access for coding agents
  • Datasets and data annotation tools

Pricing: A free plan includes 1 GB of data. Starter costs $30 and Pro $150 per month with unlimited seats and per-GB overage; Enterprise is custom with an on-premise option.

Read more about LaminarWebsiteGitHub#7 Weights & Biases alternative

Kubeflow

Kubeflow is an open source platform for deploying ML workflows on Kubernetes.

GitHub stars
16k
Last commit
1 mo ago
Latest release
redirect
Licence
Apache-2.0
Self-hosted
Yes
kubeflow.orgKubeflow homepage screenshot

Kubeflow is the Cloud Native AI platform composed of modular, open source projects that form a Kubernetes-native stack for data and AI workloads. It is designed for AI practitioners, platform administrators, and decision-makers who need modular, scalable, and extensible tools for data, AI/ML, and HPC use-cases.

The platform's mission is to bridge the Data, AI, and Cloud Native ecosystems, enabling teams to deliver more models, agents, and AI applications into production across the AI lifecycle. It is built by an open global community and aims to be the standard for Data & AI workloads on Kubernetes.

Key features

  • Runs workloads at any scale without requiring Kubernetes expertise
  • Portable across local laptops, on-premises, and any cloud
  • Manages hyperscale training jobs and high-throughput AI agents
  • Composable tools that can be mixed and matched across the AI lifecycle
  • Includes subprojects: Spark Operator, Notebooks, Trainer, Katib, and Hub
Read more about KubeflowWebsiteGitHub#8 Weights & Biases alternative

Polyaxon

Polyaxon is an open-source AI control plane for Kubernetes that schedules, runs, debugs, observes, and automates AI/ML workloads, agents, and sandboxes on your own infrastructure.

GitHub stars
3.7k
Last commit
yesterday
Licence
Apache-2.0
Self-hosted
Yes
polyaxon.comPolyaxon homepage screenshot

Polyaxon is an open-source AI control plane designed to run on Kubernetes. It enables scheduling, tracking, and governance of AI/ML workloads, including GPU training, agent sessions, and sandboxes, all within self-hosted deployments. The platform supports training, services, and sandboxes using shared specs, queues, governance, and data connections.

It provides a unified interface for training and fine-tuning models, serving inference, running notebooks, and managing interactive sessions. Polyaxon integrates with various frameworks, clients, and APIs, and offers governance features such as organizations, teams, authentication, audit logs, and policies. It is suitable for teams that need to manage AI workloads on their own clusters and storage backends.

Key features

  • Scheduling and orchestration of runs, experiments, and batch jobs
  • Training and fine-tuning with support for LoRA and SFT
  • Services for serving, inference, notebooks, TensorBoard, and ML apps
  • Sandboxes with SSH, tmux, interactive, and background modes
  • Governance with organizations, teams, authn/authz, audit logs, and policies
  • Execution controls via agents, queues, connections, and presets
Read more about PolyaxonWebsiteGitHub#9 Weights & Biases alternative

Pezzo

Open-source LLMOps platform for managing prompts, observing AI calls and delivering prompt changes, with Node.js, Python and LangChain clients.

GitHub stars
3.3k
Last commit
5 days ago
Latest release
v0.9.2
Licence
Apache-2.0
Self-hosted
Yes

Pezzo is an open-source LLMOps platform under the Apache 2.0 licence. It covers prompt management, observability and caching for applications that call language models, and lets teams change prompts and deliver them without redeploying code.

Client libraries exist for Node.js, Python and LangChain. The Python client is installed with `pip install pezzo openai` and is described as needing two lines of code to start troubleshooting and managing prompts. The README's feature table marks prompt management, observability and caching as available in all three clients.

The platform is built only on open-source components: PostgreSQL, ClickHouse, Redis and Supertokens. The README documents running the full stack locally with Docker Compose, and a development setup that needs Node.js 18+ and Docker, with a server and a console app.

Key features

  • Prompt management shared across the team
  • Observability and monitoring of AI operations
  • Clients for Node.js, Python and LangChain
  • Deploy prompt changes to any environment quickly
  • Full stack runs locally through Docker Compose
  • Built on PostgreSQL, ClickHouse, Redis and Supertokens
Read more about PezzoWebsiteGitHub#10 Weights & Biases alternative

4 more Weights & Biases alternatives

Commercial Weights & Biases alternatives

Hosted, closed-source products that compete with Weights & Biases. Showing 12 of 25.

See every Weights & Biases alternative on its tool page

Weights & Biases alternatives: questions

What is the best open-source alternative to Weights & Biases?
Langfuse is the top-ranked open-source alternative to Weights & Biases on Enlisted: An open-source LLM engineering platform for tracing, evaluating and improving AI applications, with prompt management, datasets and a playground, self-hosted or cloud. Other strong options are TraceRoot, Opik, Arize Phoenix and MLflow.
What are the best commercial alternatives to Weights & Biases?
Dataiku, DataRobot, Amazon SageMaker, Domino Data Lab and H2O.ai are among the hosted, closed-source alternatives to Weights & Biases listed on Enlisted, which lists 25 in total. Compare their pricing and features on each tool page.
Are these Weights & Biases alternatives free?
All 14 are open source, so the code is free to use under its licence, and all of them can be self-hosted on your own server or computer. 7 also offer a paid or managed cloud version if you'd rather not host it yourself.
How is this list of Weights & Biases alternatives ranked?
By a score built from GitHub stars, star growth over the last 30 days and how recently the code changed. 9 of these projects shipped code in the last 30 days. Data is refreshed daily, and nobody can pay to move up.

People also look for alternatives to…

View all