Score - Subnet 44

@MaxScore 🚗

less than 24hrs after launching the new vehicle detection “skill” challenge on sn44… the incentive mechanism clearly works as expected next: the private track

Computer vision is one of the most commercially valuable AI capabilities right now. We found out how @WeBuildScore is making powerful, specialized vision models that are small enough to run on a CPU, in collaboration with Bittensor miners. Find out how they’re turning crowdsourced computer vision into a real business on TWiST

The @manakoai desktop agent is born... No cloud pipelines. No egress tax. No latency games. Download → it finds your cameras → deploy vision agents locally → inference runs on your hardware. Only evidence + telemetry hit the cloud. We're moving vision compute to where it belongs: the edge.

From live feed to your phone in seconds.

We’re live on @twistartups @MaxScore is speaking on the future of vision AI and Score’s & @manakoai approach to programmable vision.

This Bittensor Subnet Could Cut Drug Discovery Costs in Half x.com/i/broadcasts/1…

It's an all $TAO special today, with Tom and Will Blears of @Bitcast_network (Subnet 93) PLUS @MaxScore of @WeBuildScore (Subnet 44) also joining. Follow all of our plans on the live docket: thisweekinstartups.com/docket

@tm0klc 👁️

In just a week, our PersonDetect element improved from 41.9% to 78.4% mAP@50 while staying under 30 MB. That's an 87% relative lift over baseline. No manual retraining cycle, the subnet incentivises miners to compete on a clear benchmark, and the model improves because the structure rewards improvement. The next milestone is 85% mAP@50, and the current trajectory suggests we'll likely exceed it. VehicleDetect is next, and we expect to see the same kind of rapid improvement curve.

The New Detect Vehicle Task is Live! #3 of many. Can’t Stop, Won’t Stop.

.@webuildscore on @twistartups tomorrow (Mar 25) stay tuned chaps

Follow @stillcorecap and @markjeffrey if you want to follow the $tao movement — he’s one of the few professionals evaluating and investing in the subnets

Yuma's Large Cap Subnet Fund is up 76% YTD on a USD basis; 46% higher than $TAO/USD growth over the same period. The growth is driven by targeted exposure to Bittensor's largest subnets like @tplr_ai (SN3), @TargonCompute (SN4), and @webuildscore (SN44), among others.

Instead of a product department, imagine you make a list of TASKS. AI devs, agents and raw GPU from all over earth compete to complete your task the best and fastest. A Bitcoin-like chain rewards the winner with newly-minted tokens. That's Bittensor.

NEW TASK ON SN44 Our first vehicle detection challenge goes live tomorrow. Car, bus, truck, motorcycle…on CCTV footage. MINER STARTER PACKs are now baked directly into the console (link in first comment) Requirements: ∙Public track ∙50ms latency ∙30fps ∙<30mb model ∙Baseline 40.7% mAP → target 90% As @tm0klc put it: “expect the frequency of tasks to increase in the coming weeks” Happy Monday and Happy Mining!

@MaxScore 👀

btw that was not about eyeball

Score x @Eyeballfootball We're partnering with Eyeball to bring Vision AI to youth football, turning match footage into real-time player development intelligence that helps identify and grow talent from grassroots pitches to elite academies.

@tylerdurdeth 👁️

This is incredible. Every business operating CCTV is gonna want to use this tool to turn thousands of hours of passive recordings into useful information that can triggers action. IMO, @webuildscore is about to have a Templar moment. Built on Bittensor ($TAO)