Bittensor subnet
Witness
A competition to build open AI models that describe what's seen and heard in videos.
ResearchVideo
τ
υ
Can I use it today?
Visit their site ↗Not as a customer. It's a contest, not a product.
Research. Runs a competition between miners, with nothing to sign up for.
TAO in its pool6.5K#66 of 128 subnets
Share of new TAOUnder 0.1%Next to nothing right now
Its tokenυ0.0024 TAO each
In this slot sinceDec 20231,017 days running
What you can do with it
- View the current best video model and its scores on the dashboard
Things to know
No app, API or pricing for non-miners is mentioned.
TAO in its pool, month by month
Similar subnets
SN44ScoreBuild, train and run AI that understands camera video, like detecting, tracking and counting objects.SN53engyPay-as-you-go access to top open AI models for text, image reading and video, with no data kept.SN117everyframe.studioAn AI video studio that gives you Veo, Kling, Seedance and MiniMax on one credit balance.SN64ChutesPay-as-you-go access to open AI models for chat, images, video, speech and music.
Sources
Checked October 13, 2026. Pool and emission numbers read from the chain October 13, 2026.