Skip to main content
Kapso is a self-improving software factory. State an objective and it runs a campaign: it designs candidate solutions, has coding agents implement them, measures how far each one is from the objective, and keeps refining the closest until the objective is met. The result ships to your infrastructure. The factory improves with use. When a campaign ends, Kapso studies its own work: which ideas closed the distance to the objective, which did not, and under what conditions. Each finding is kept as a lesson with the evidence that earned it, and a lesson stays trusted only as long as it keeps holding up. Kapso also reads outside your walls, repositories and papers, and folds what it finds into the same knowledge hub. Every new campaign begins from that hub, so it starts with what earlier work already established, about the problem and about your systems. Kapso is an open-source Python framework by Leeroo. The source lives at github.com/Leeroo-AI/kapso and the package installs with pip install leeroo-kapso.

The four pillars

Evolve

.evolve() runs iterative experiments to build software for a goal, using tree search, coding agents and knowledge-graph context. When a session needs a credential or a file only you can provide, the campaign asks and pauses until you reply.

Learn

Two memories. .learn() mines your own finished campaigns into evidence-priced lesson cards. .learn_knowledge() ingests repositories and research into the knowledge graph.

Research

.research() runs deep web research and returns structured findings you can feed into the knowledge graph or pass as campaign context.

Deploy

.deploy() turns a solution into running software, locally, in Docker, or on a cloud platform such as Modal.

What a full run looks like

learn() mines your own finished campaigns; learn_knowledge() ingests outside sources. They are different memories.

Results

IOAI 2026

Competing fully autonomously in the AI Model Track, Kapso scored 536.07 — above every one of the 471 human contestants — and took the IOAI² Grand Master Trophy, top 3 among all AI systems entered.

RelBench

Passes KumoRFM-v2 on outcome prediction and forecasting, and the best reported results on recommendations, on Stanford’s benchmark for predictive ML over relational data.

MLE-Bench

Top ranking among open-source systems on Kaggle machine-learning competitions.

ALE-Bench

Top ranking on long-horizon algorithmic discovery problems from AtCoder.
Bar chart of three RelBench panels: outcome prediction, Kapso 81.2 AUROC against KumoRFM-v2's 79.6; forecasting, Kapso 0.2476 NMAE against 0.2912, lower being better; recommendations, Kapso 18.4 MAP against the best reported 15.7. Each panel is drawn from its own truncated axis

Kapso against KumoRFM-v2 on RelBench outcome prediction, forecasting and recommendations

The technical report describes the system in full.

What is inside

Leeroopedia MCP

Kapso connects to Leeroopedia, a curated knowledge base of ML and AI frameworks. Agents query it during ideation and implementation to look up APIs, build plans and diagnose failures. See Optional: Leeroopedia MCP.

Start here

Installation

The CLIs, the package, and kapso doctor

Quickstart

Your first campaign

CLI reference

Every command and flag
Related pages: Evolve · Learn · Research · Deploy · IOAI 2026 · RelBench · MLE-Bench · ALE-Bench · Installation · Quickstart · CLI reference Kapso is an open-source framework by Leeroo that builds software toward measurable goals through experiment campaigns. Source code: github.com/Leeroo-AI/kapso · Install: pip install leeroo-kapso · Every page as plain text: llms.txt.