Langfuse alternatives (2026)
Langfuse is a evals & observability tool: Open-source LLM observability, tracing and evals. Pricing, as the maker states it: Open source + paid cloud. Its GitHub repo has 36,432 stars. We found 42 tools in the directory that do overlapping work (evaluating agents, monitoring agents in production); 23 of them are open source and 10 list a free or freemium plan. Below are the 10 closest, ranked by how many tasks and sub-tasks they share with Langfuse, then by GitHub stars.
Updated
Quick answer: the top picks on this list, ranked by GitHub stars and listing maturity, are MLflow, Opik and Arize Phoenix, out of 11 tools listed below. At a glance: 5 open-source, 10 free or freemium and 5 local or self-hosted across 11 link-checked tools (last verified ).
Verified on : every tool link on this page was checked by our daily link check. No paid placements.
Ranked comparison (top 10)
| # | Tool | Pricing | Best for | Open source | Runs locally | GitHub stars |
|---|---|---|---|---|---|---|
| 1 |  MLflow | Open source + cloud | automate workflows, monitor & evaluate agents | Yes | Self-hosted | 28,526 |
| 2 |  Opik | Open source + cloud | automate workflows, monitor & evaluate agents | Yes | Self-hosted | 22,902 |
| 3 |  Arize Phoenix | Source-available + cloud | automate workflows, monitor & evaluate agents | No | 11,995 | |
| 4 |  Agenta | Open source + cloud | monitor & evaluate agents | Yes | Self-hosted | 4,859 |
| 5 |  Future AGI | Open source + cloud | monitor & evaluate agents | Yes | Self-hosted | 2,199 |
| 6 |  W&B Weave | Freemium | build an agent, monitor & evaluate agents | No | 1,142 | |
| 7 |  LangSmith | Freemium | automate workflows, monitor & evaluate agents | No | 17 | |
| 8 |  Arize AI | Freemium | monitor & evaluate agents | No | ||
| 9 |  Braintrust | Freemium | monitor & evaluate agents | No | ||
| 10 |  Confident AI | Freemium | monitor & evaluate agents | No |
- MLflow (Open source + cloud): Open-source platform for tracing, evaluating and managing agents, LLM apps and ML models. Consider it because it does the same job with a different approach (open source + cloud).
- Opik (Open source + cloud): Comet's open-source platform for tracing, evaluating and monitoring LLM apps, RAG systems and agent workflows, self-hosted or in the cloud. Consider it because it also covers RAG and documents and workflow automation.
- Arize Phoenix (Source-available + cloud): Open-source tracing and evaluation for LLM apps. Consider it because it does the same job with a different approach (source-available + cloud).
- Agenta (Open source + cloud): Open-source LLMOps platform for prompt management, evaluation and observability. Consider it because it does the same job with a different approach (open source + cloud).
- Future AGI (Open source + cloud): Open-source platform for evaluating, tracing, simulating and guarding LLM and AI agent applications, self-hostable or cloud. Consider it because it does the same job with a different approach (open source + cloud).
- W&B Weave (Freemium): Weights & Biases toolkit for tracing and evaluating agents. Consider it because it also covers agent frameworks and SDKs.
- LangSmith (Freemium): Tracing, evals and monitoring for LLM apps and agents. Consider it because it does the same job with a different approach (freemium). Langfuse vs LangSmith →
- Arize AI (Freemium): Agent observability and evaluation platform with tracing, evals and experiments for improving AI agents. Consider it because it does the same job with a different approach (freemium).
- Braintrust (Freemium): Evals and observability platform for AI products. Consider it because it does the same job with a different approach (freemium).
- Confident AI (Freemium): Enterprise platform for AI evals and observability that gives teams one standard for measuring and monitoring AI quality. Consider it because it does the same job with a different approach (freemium).
Langfuse vs alternatives at a glance
| Tool | Pricing | Open source / licence | GitHub stars | Good for |
|---|---|---|---|---|
 Langfuse | Open source + cloud | MIT, except the ee/ directories (commercial licence) | 36,432 | evaluating agents, monitoring agents in production |
| MLflow | Open source + cloud | Apache License 2.0 | 28,526 | automating workflows, evaluating agents |
| Opik | Open source + cloud | Apache License 2.0 | 22,902 | automating workflows, evaluating agents |
| Arize Phoenix | Source-available + cloud | Elastic License 2.0 (source-available) | 11,995 | automating workflows, evaluating agents |
| Agenta | Open source + cloud | MIT License, except ee/ directories (separate licence) | 4,859 | evaluating agents, monitoring agents in production |
| Future AGI | Open source + cloud | Yes | 2,199 | evaluating agents, monitoring agents in production |
| W&B Weave | Freemium | Apache License 2.0 | 1,142 | building AI agents, evaluating agents |
| LangSmith | Freemium | MIT | 17 | automating workflows, evaluating agents |
| Arize AI | Freemium | No | — | evaluating agents, monitoring agents in production |
| Braintrust | Freemium | No | — | evaluating agents, monitoring agents in production |
| Confident AI | Freemium | No | — | evaluating agents, monitoring agents in production |
Written comparisons: Langfuse vs LangSmith
More: Langfuse listing · All evals & observability tools · All alternatives pages
Related searches: Open-source LLM observability and eval tools
Alternatives are picked from directory data only: same category plus shared tasks and sub-tasks. Pricing is as each maker states it; GitHub stars as recorded on each listing.