multivon-eval is an open-source Python framework for evaluating large language models, agents, and retrieval-augmented generation pipelines. It provides statistical rigor, confidence intervals, and agent trace analysis, supporting local-first workflows for ML engineers and researchers.
Multivon is a LLM eval & observability product. It focuses on measuring and validating the performance of LLMs, agents, and RAG pipelines with robust statistical methods. It is built as an open-source project for machine learning engineers. Multivon is open source under the Apache-2.0 license. It runs on the command line.
It is developed by Multivon, and the product first shipped in 2026. Development happens publicly on GitHub with 227 commits in the last 90 days. PulseGate's similarity index finds few close equivalents — Multivon occupies a relatively distinct niche. Key capabilities include statistical evaluation, confidence intervals, and agent trace analysis.
Latest indexed changes and source events
multivon.ai discovered by the PulseGate indexer
Other apps tracked under the same category.
