---
title: "Zeluna — Be your own frontier"
description: "Zeluna is the autoresearch agent for AI products. It builds the environment your product lives in, runs the research against it, and ships what moved."
url: https://zeluna.ai/
---

# Be your own frontier.

Zeluna is the autoresearch agent for AI products. It builds the environment your product lives in, runs the research against it, and ships what moved.

[Request access](https://cal.com/caua-veiga-jn24rz/30min)

## Your own lab, running one loop.

Every lap starts from what your product actually did, and ends with a model that does it better.

1. **Benchmark.** A private benchmark of your own work, built from your traces and graded on what happened next (the ticket reopened, the refund never filed), not on an LLM judge's opinion.
2. **Research.** An agent runs the experiments: models, prompts, tools and routing, tested against your benchmark. It keeps what held, proposes it for your review, and writes up why.
3. **Train.** When the benchmark is rich enough, it becomes the reward: a smaller open model, post-trained on your outcomes. The weights are yours.
4. **Ship, and again.** Production traces flow back into the benchmark, and every new model is tested against it. Each lap starts higher.

You connect a tracing tool and the table where the outcome lands. Nothing else.

## Research

- **Frontier pass rate from an open model** (Frontier, coming soon). An open model against the frontier on τ³ banking, rung by rung, with the cost of each.
- **Do judges agree with what happened?** (Evaluation, coming soon). Models ranked twice on the same cases: by LLM judge and by recorded outcome.
- **SFT vs RL on business traces** (Post-training, coming soon). Same traces, same reward, a recorded outcome. Which post-training method wins, and by how much?
- **Fidelity of synthesised environments** (Environments, coming soon). Executability is not correspondence. A metric for whether the synthesised world predicts production.

## Questions

### What does it take to start?

Two connections: your production traces, and read access to wherever the outcome is recorded, such as a ticket reopening or a refund being filed. No labelling, and no new pipeline.

### Do I need my own GPUs or infrastructure?

No. The lab runs on ours: building the benchmark, the experiments, training and serving. Your team connects the data and reviews what the lab proposes.

### How is this different from an eval or fine-tuning platform?

Those start from a dataset or a judge you have already built. We start from what your product actually did, grade it on what happened next, and keep the loop running as new traffic comes in.

### Who owns what comes out?

You do. The benchmark and the weights of every model we train are yours to export and run wherever you like. Or leave them with us: we store them, serve the model behind an endpoint, and keep the lab running on it.

### How do I get access?

We're in a closed beta with a small number of teams whose product records an outcome to learn from. Request access and we'll look at your use case together, and tell you plainly whether it's a fit now.

## Be the frontier.

[Request access](https://cal.com/caua-veiga-jn24rz/30min) · [GitHub](https://github.com/Zeluna) · [For AI agents: llms.txt](https://zeluna.ai/llms.txt)

---

© 2026 Zeluna. The product screens on https://zeluna.ai show a fictional company (Acme) with invented numbers; they are not results.