llms.txt
Compare with other sites' llms.txt files →# Oumi
> Oumi is the AI that builds your AI. Train models and agents. Run inference. Improve continuously. Fully automated, transparent, and under your control.
A specialized model trained on your task is typically 10x to 100x smaller than a
frontier model, yet matches or beats it on that task at a fraction of the
inference cost. You own the weights, the data and the recipes, and you can run
them in Oumi's cloud or your own infrastructure. Oumi is also an open-source
project (Apache 2.0).
## Product
- [Train & improve](https://oumi.ai/platform/train): Train, evaluate and continuously improve specialized models and agents in one platform, with FAiE automating the AI engineering work under your control.
- [Inference](https://oumi.ai/platform/inference): Run production inference on Oumi-managed GPUs, in your private VPC, or on-premises, and feed production signals back into the next improvement cycle.
- [Infrastructure & developer tools](https://oumi.ai/platform/infrastructure): Use Oumi from the CLI, Python API, and AI coding tools, with the security and infrastructure controls required for production AI.
- [Oumi platform](https://platform.oumi.ai/): The product itself — build, evaluate, train and deploy specialized models.
- [Sign in](https://platform.oumi.ai/signin): Start for free with up to $50 in credits.
- [Documentation](https://docs.oumi.ai): Platform and open-source library docs.
- [Pricing](https://oumi.ai/pricing): Plans and what each tier includes.
- [Rent vs Own Calculator](https://oumi.ai/calculator): Compare the cost of a rented frontier API against an owned specialized model.
- [Demos](https://oumi.ai/demos): Interactive demos and walkthroughs of what you can build.
## Evidence
- [Case studies](https://oumi.ai/case-studies): How enterprises use Oumi to beat frontier APIs on accuracy, cost and data control.
- [Atlas of AI](https://oumi.ai/atlas): 580 Specialist Models, one per business decision, each with a guide to building and owning it.
- [Oumipedia](https://oumi.ai/oumipedia): Oumipedia explains fine-tuning and open LLMs in plain language: what fine-tuning is, RAG vs. fine-tuning, LoRA vs. QLoRA, and open model profiles.
- [Resources](https://oumi.ai/resources): Oumi resources in one library: fine-tuning guides, eBooks, webinars, demos, case studies, product launches, press and podcasts. Filter by type.
- [What Is Fine-Tuning?](https://oumi.ai/oumipedia/what-is-fine-tuning): Fine-tuning trains an existing LLM on your own examples so it does one task better. Learn how LLM fine-tuning works, when to use it and its limits.
- [What Is Custom Fine-Tuning?](https://oumi.ai/oumipedia/what-is-custom-fine-tuning): Custom fine-tuning trains an open model on your company's data for one task. Learn what it takes, what it costs and how it compares to prompting.
- [RAG vs. Fine-Tuning: Which Should You Use?](https://oumi.ai/oumipedia/rag-vs-fine-tuning): RAG looks facts up at answer time. Fine-tuning teaches a model a behavior. See when to use RAG vs. fine-tuning, with a decision matrix and an example.
- [LoRA vs. QLoRA vs. Full Fine-Tuning: How to Choose](https://oumi.ai/oumipedia/lora-vs-qlora-vs-full-fine-tuning): LoRA vs. QLoRA vs. full fine-tuning: how each method works, how to choose one, and what each choice means when you deploy the model.
- [Fine-Tuning Methods Guide: Which One Is Right for You?](https://oumi.ai/resources/guides/fine-tuning-methods): Compare fine-tuning methods: SFT, LoRA, QLoRA, DPO and GRPO. Learn what each one changes in the model, what it needs from your data and when to use it.
- [Blog](https://oumi.ai/blog): Insights on custom models, open-source ML, and specialized intelligence.
- [Newsroom](https://oumi.ai/newsroom): Press coverage, announcements and research highlights.
## Company
- [About Oumi](https://oumi.ai/company): Mission, team and values.
- [Careers](https://oumi.ai/careers): Open roles.
- [Partners](https://oumi.ai/partners): AWS, Databricks and Nebius, the platforms Oumi works with.
- [Contact](https://oumi.ai/contact): Reach the team.
- [Events](https://oumi.ai/events): Where Oumi is showing up next, plus on-demand sessions.
## Community
- [GitHub](https://github.com/oumi-ai/oumi): The open-source Oumi library, Apache 2.0.
- [Discord](https://discord.gg/oumi): Practitioners building with Oumi, in the open.
- [Substack](https://oumiai.substack.com/): Long-form writing from the team.
## Optional
- [Job board](https://jobs.ashbyhq.com/oumi): Full listing behind the careers page.
- [Sitemap](https://oumi.ai/sitemap.xml): Every indexable URL, including individual blog posts.