@CosineAI: An early version of Lumen Sovereign – our AI model trained entirely in the UK – is running locally on an Apple Mac Stud…
Summary
CosineAI announces an early version of Lumen Sovereign, a UK-trained AI model running locally on an Apple Mac Studio, with plans to scale to a 1 trillion-parameter model.
View Cached Full Text
Cached at: 09/04/26, 12:15 AM
An early version of Lumen Sovereign – our AI model trained entirely in the UK – is running locally on an Apple Mac Studio in our office.
This 200B baseline is just the beginning.
We’re now breaking ground on a 1 trillion-parameter model, actively synthesising RL data from real-world use cases for complex tasks provided by our coalition and public RFI.
📰 Keep up to date with Lumen Sovereign by joining our newsletter: https://tally.so/r/1AzGq1
Get the Latest on Lumen Sovereign
Source: https://tally.so/r/1AzGq1
Page 1 of 2Join the newsletter to receive the latest news on Britain’s first fully sovereign frontier AI model.
Subscribe to receive:
– Development Updates: track the model’s progress and capabilities.
– Exclusive Previews: be the first to see new demos, videos, and model checkpoints.
– Industry Insights: key news and commentary on the future of secure, air-gapped AI.
Name
Company E-mail Address
Similar Articles
@DeRonin_: My current local AI setup: - 2x DGX Spark linked (256gb) > GLM 5.2 @ 2bit, reasoning + agent loops - Mac Studio M3 Ultr…
A user describes their fully local AI stack using multiple hardware devices running Chinese models like GLM, Qwen, and Kimi, claiming 87% cost savings compared to frontier models like GPT-5.5 and Opus 4.8, while noting plans to self-host video generation.
@Saboo_Shubham_: This is the FUTURE. Local models running on your phone with borrowed compute from your MacBook. 100% free and works wit…
LM Studio enables running local open-source AI models on your phone by borrowing compute from your MacBook, 100% free.
My local model setup on an M4 Pro Mac Mini
The author describes their local AI model setup on an M4 Pro Mac mini, using models like Qwen and Gemma with tools such as oMLX and Tailscale to achieve data privacy, cost predictability, and offline capability.
@rohanpaul_ai: atomic[.]chat (a desktop app that runs LLMs locally) ran a very revealing comparison for local AI agents, on a MacBook …
Liquid's LFM2.5-8B-A1B outperformed OpenAI's gpt-oss-20b on a tool-calling benchmark when run locally on a MacBook Pro, completing all required tool calls in half the time while using less memory.
@DanKornas: Fine-tuning local LLMs shouldn’t require renting a cloud GPU. Silicon Studio is an open-source desktop app for local LL…
Silicon Studio is an open-source desktop app that enables local LLM fine-tuning and inference on Apple Silicon Macs using MLX, with features for data preparation, model management, and visual configuration.