Open vs closed models — when to choose which?

Open for control, privacy, and cost at scale; closed when you want the absolute frontier and don't want to run infrastructure.

What hardware do I need to self-host?

It depends on model size and quantization — small models run on a single consumer GPU; large ones need serious accelerators or a hosted inference provider.

Do Oort prompts work on open models?

Generally yes — well-structured prompts travel across models, and Oort is BYOK so you can point them at any open model you host or call.

Open-Source LLMs

Open-weight model releases, licenses, and self-hosting news — the fast-moving frontier of models you can run yourself.

What to know

Open-weight models — Llama, Mistral, Qwen, Gemma, DeepSeek and more — let you self-host, fine-tune, and avoid per-token fees. The space moves weekly: new releases, license changes, and quantized community builds. This hub tracks what's actually worth running.

Latest Open-Source LLMs news

No headlines yet

We're tracking Open-Source LLMs — fresh coverage lands here as it breaks.

Open-Source LLMs — FAQ

Open vs closed models — when to choose which?: Open for control, privacy, and cost at scale; closed when you want the absolute frontier and don't want to run infrastructure.
What hardware do I need to self-host?: It depends on model size and quantization — small models run on a single consumer GPU; large ones need serious accelerators or a hosted inference provider.
Do Oort prompts work on open models?: Generally yes — well-structured prompts travel across models, and Oort is BYOK so you can point them at any open model you host or call.