Ollama Can Speak Now: Native Text-to-Speech with Phonon
Ollama added native speech generation on 24 October. How to get it talking, which checkpoint to pull, the licence trap, and what still returns a 400.
Read article
All articles
Local LLMs
3 articlesRun AI models on your own hardware. Guides for Ollama, LM Studio, and self-hosted inference — no cloud, no API costs.
Running a language model on your own machine changes the economics of everything built on top of it. There is no per-token bill, no rate limit, no request leaving your network, and no vendor deprecating the model you built a workflow around. The trade is that you own the setup: picking a model that fits your VRAM, quantizing it so it actually runs, and accepting that a 7B model on a laptop is not going to match a frontier API on reasoning-heavy work.
These guides cover the practical end of that — installing and running Ollama, picking between model families and quantization levels, putting a usable interface in front of a local model, and deciding which jobs are worth keeping local versus sending to an API. Every benchmark and hardware note here comes from a Ryzen 7 5800HS with 16GB of RAM and a GTX 1650, which is deliberately unremarkable hardware. If a workflow runs there, it runs on most things.
Start with the Ollama guide if you have never run a local model, then move to Open WebUI once you want something you can actually hand to someone else.
Do You Actually Need a Local LLM? A Business Guide to Keeping Client Data Private
Your paid AI subscriptions already promise not to train on your data. When a local LLM for business is still the right answer, and what it really costs.
Read article
The Complete Guide to Running Local LLMs with Ollama in 2026
Everything you need to set up Ollama, pick the right model, and start running AI locally on your own hardware — no cloud, no API costs, full privacy.
Read article