Oumipedia is Oumi’s reference for building specialized AI: plain explanations of the core concepts, and side-by-side comparisons of methods and models.

Where are model profiles?

Here in Oumipedia. Each open-weight model family Oumi supports has a profile, next to the comparisons between them.

Oumipedia

Your reference for building specialized AI: what the concepts mean and how the options compare.

13 of 13 entries

Explainer
What Is Fine-Tuning?
3 min read
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.
Read: What Is Fine-Tuning?
Explainer
What Is Custom Fine-Tuning?
5 min read
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.
Read: What Is Custom Fine-Tuning?
Comparison
RAG vs. Fine-Tuning: Which Should You Use?
10 min read
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.
Read: RAG vs. Fine-Tuning: Which Should You Use?
Comparison
LoRA vs. QLoRA vs. Full Fine-Tuning: How to Choose
9 min read
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.
Read: LoRA vs. QLoRA vs. Full Fine-Tuning: How to Choose
Explainer
What is Qwen?
Models · Profile · Alibaba Cloud
Qwen is Alibaba Cloud’s open-weight model family. It spans small models that are practical to fine-tune through larger mixture-of-experts releases.
Read: What is Qwen?
Explainer
What is DeepSeek?
Models · Profile · DeepSeek
DeepSeek publishes open-weight language models, including large mixture-of-experts systems. The family is often shortlisted where reasoning and coding matter.
Read: What is DeepSeek?
Explainer
What is Llama?
Models · Profile · Meta
Llama is Meta’s open-weight model family. Its wide ecosystem makes it a familiar baseline for teams evaluating or fine-tuning their first custom model.
Read: What is Llama?
Explainer
What is Mistral?
Models · Profile · Mistral AI
Mistral is a European model family spanning open-weight and commercial releases. The right starting point depends on the exact release and deployment constraints.
Read: What is Mistral?
Explainer
What is Gemma?
Models · Profile · Google
Gemma is Google’s open-weight model family. It includes compact checkpoints that can be useful when hardware limits and deployment footprint drive the decision.
Read: What is Gemma?
Comparison
Qwen vs. DeepSeek
Models · Comparison
Are you optimizing for a smaller model you can specialize, or a larger reasoning and coding candidate?
Read: Qwen vs. DeepSeek
Comparison
Qwen vs. Llama
Models · Comparison
Does your task evaluation favor the Qwen family or the Llama ecosystem and license fit?
Read: Qwen vs. Llama
Comparison
What is the best open-source LLM?
Models · Selection guide
There is no universal winner. Pick a model family, then evaluate the exact checkpoint against your real task, output contract, deployment budget, and license constraints.
Read: What is the best open-source LLM?

Questions about Oumipedia

What is Oumipedia?
Oumipedia is Oumi’s reference for building specialized AI: plain explanations of the core concepts, and side-by-side comparisons of methods and models.
Where are model profiles?
Here in Oumipedia. Each open-weight model family Oumi supports has a profile, next to the comparisons between them.