Strictly a research tool for neuroscientists and AI researchers modelling brain responses; free, but non-commercial use only.
Kenning's take
What it does
TRIBE v2 is a research model from Meta that predicts high-resolution fMRI brain activity in response to images, audio, video and text. It was trained on recordings from more than 700 volunteers and can make predictions for new people, languages and tasks it has not seen. Meta released the paper, weights and code for non-commercial use, plus an online demo.
- Predicts fMRI brain responses to sight, sound and language
- Trained on data from 700+ volunteers
- Zero-shot predictions for new subjects and tasks
- Weights on Hugging Face and code on GitHub
- Interactive online demo
Who it suits. Neuroscience and AI researchers who want to test ideas about the brain before running human studies.
Key facts
- Category
- AI models
- Made by
- Meta
- Released
- 26 March 2026
- How to get it
- free research release: weights on Hugging Face, code on GitHub, demo on Meta's site; CC BY-NC licence
- Works on
- open-weights, Web
- In the UK
- Yes
- Free plan
- free weights and code, non-commercial licence
FAQ
When was TRIBE v2 released?
Meta announced TRIBE v2 on 26 March 2026.
What does TRIBE v2 do?
It predicts how the brain responds, as measured by fMRI, to images, sounds, video and text.
Can TRIBE v2 be used commercially?
No. Meta shared the paper, weights and code under a CC BY-NC licence, which allows non-commercial use only.
Facts from the maker's own pages and the sources listed. Checked 2026-10-11.
Similar tools
ACE-Step 1.5
Open-source music generation model that writes full songs with lyrics and runs on ordinary consumer GPUs.
AlphaGenome Atlas
Google DeepMind database of AI-predicted effects for every possible single-letter change in human DNA.
Amazon Nova 2
Amazon's second family of Nova AI models, covering reasoning, speech conversation and combined text-and-image generation.
Cisco Antares
Cisco's small open-weight models that find where known security vulnerabilities sit in a codebase, running locally.