TAOQuant
Quant Subnet Analytics & Risk Matrix for Bittensor

Bittensor analytics
TAOQuant
Deterministic subnet intelligence, risk scoring, and APY stability analysis in one auditable workspace.
IRA Model Parameters
Adjust the weights of the Adjusted Risk Index equation. Weights always normalize to 100%.
Subnet Analytics
Data source: Local mock dataset (no requests spent)
SN1Text Prompting | 14.2% | 0.18 | 0.22 | 9% | 83Low Risk |
SN8Time-Series Prediction | 9.40% | 0.12 | 0.29 | 7% | 83Low Risk |
SN21FileTAO Storage | 12.1% | 0.25 | 0.33 | 14% | 75Medium |
SN2Machine Translation | 11.7% | 0.31 | 0.41 | 18% | 69Medium |
SN5Image Generation | 18.9% | 0.44 | 0.38 | 21% | 64Medium |
SN27Compute Subnet | 16.6% | 0.39 | 0.48 | 26% | 61Medium |
SN4Multi Modality | 22.5% | 0.62 | 0.55 | 34% | 47High Risk |
SN18Cortex.t | 27.8% | 0.78 | 0.67 | 46% | 34High Risk |
Model Math & Audit
The IRA is a fully transparent, reproducible weighted score. No black boxes.
IRA = 100 × (
w1 · stability +
w2 · decentralization +
w3 · efficiency
)
// each pillar is clamped to [0, 1] — never negative
stability = max(0, 1 − CV)
decentralization = clamp(1 − HHI)
efficiency = max(0, 1 − churn / 100)
// churn is a percentage (0..100); scale clamped to [0,1]
// constraint
w1 + w2 + w3 = 1.0Statistical pillars
APY Stability
Rewards consistent yield. Derived from 1 − normalized Coefficient of Variation.
HHI Decentralization
Rewards distributed staking. Derived from 1 − Herfindahl-Hirschman Index.
Efficiency / Churn
Rewards network health. Derived from 1 − normalized miner churn rate.
Every score above is computed client-side from the values in the table. Change a weight and re-audit instantly.
Support TAOQuant Development
Free and open-source tool for the TAO community. If this model helped you, support the project maintenance:
5D4tdPN57Qc379hJMdStBdNmLxynpQBGv8KUa5PBhNpf8zVY