Neuron registration trends
The alpha price in τ, day by day, and what moved it each way.
Log change over the last 30 days −3.01%: protocol buys +0.06%, root sells −0.66%, users net −1.45%, cross-subnet −0.96%, unattributed −0.0004%.
The chart follows the alpha price in τ. Each period starts at its opening price and steps through the mechanisms below, in a fixed order, to its closing price: a step that rises pushed the price up, one that falls pushed it down, and its height is how far it moved the price. A thin line carries each close into the next period’s open; it is dashed where the two differ. The faint candle behind each period is the market’s own open, high, low and close, for context.
With many periods on screen (90 days, or hours), the steps would be too thin to read, so the chart draws the price as a line instead and colours each period by the mechanism that moved it most. Hover a period for the full step-by-step breakdown.
Not here, because they do not move the spot price: the per-block TAO injection into the pool, and burns.
All figures are log changes, not simple percentage changes, so the parts add up exactly to the total; on big moves the two differ (a 10% log fall is a 9.5% price fall).
2026-11-05 UTC, as of ≈23:43 UTC (block 9,443,207)
The day isn’t over yet, so these figures keep changing. They are in no daily total, average or rank.
Market microstructure researchers and algorithmic trading developers struggle to access realistic, high-fidelity order-book datasets and simulation environments. Existing approaches require either expensive proprietary data feeds or simplified models that miss the order-level details necessary to develop and validate sophisticated trading strategies. Distributed networks of independent validators can collectively generate diverse, Level-3 order-book data at scale while allowing heterogeneous trading agents to interact in a shared simulated market — creating a common research substrate without central coordination.
SN79 (MVTRX) implements this as a decentralized subnet where validators run a C++ agent-based simulator maintaining multiple simultaneous orderbooks with configurable background agent models, miners deploy trading strategies that respond to full order-book state broadcasts, and the network scores miner performance on making (spread capture) and skill (risk-adjusted profit). All participants are economically incentivized: validators earn for maintaining fair simulation, miners earn for intelligent risk-managed trading, and both contribute to a shared research-quality dataset.
A secondary GenTRX mechanism overlays distributed training of a shared order-book generative model (12M-parameter transformer) on top of the simulation data. Miners compute gradients locally each round on their assigned data slice, upload to personal S3 buckets (discovered via on-chain commitments), validators score gradients against held-out data, and the canonical aggregator publishes the winning checkpoint on-chain. This allows the network to train models at scale without centralized infrastructure.
Auto-generated from the GitHub repository and may not reflect the latest changes.
MVTRX (SN79) is a Bittensor subnet for decentralised market research and AI model training, comprising τaos (agent-based trading simulation), GenTRX (distributed transformer training on order-book data), and an off-chain MVTRX Exchange. Miners develop trading strategies in simulated markets while optionally training a shared generative model; validators run the C++ simulator, score miners, and aggregate gradients.
The subnet comprises three integrated components. The C++ simulator (deployed on validators) maintains multiple simulated orderbooks and background agents, processing miner instructions and generating Level-3 market-by-order data. The Python validator acts as a proxy, receiving simulation state updates, broadcasting them to miners via dendrite, collecting miner responses, submitting instructions back to the simulator, and calculating scores.
Optionally, a gradient server (sidecar process on validators) replays simulation ticks into training parquets, collects miner gradients from per-miner S3 buckets (discovered via chain commitments), double-scores them against held-out data, aggregates locally, and publishes proposals; the canonical aggregator (uid-0) evaluates all proposals and publishes the winning checkpoint on-chain. Miners develop trading agents that respond to state broadcasts with orders; GenTRX-enabled miners also train the shared model in-process (or via HTTP to a training service) and upload compressed gradients each round. All bucket commitments (S3 read credentials) are written to chain so participants discover each other without pre-configuration.
Trading pool: scoring on making (spread capture from orders still held shortly after fill) and skill (risk-adjusted profit with market-drift effect on inventory removed). GenTRX training pool: scoring on gradient quality against held-out order-book data, with double scoring (own-data + validation set) and overfit penalty when own-data loss exceeds held-out loss by >3×. Weights are set each round as a blend of trading-pool and training-pool scores, with training contribution scaled by active miner participation.
Inactive miners require minimum trading volume to receive full allocation.
The sections above were auto-generated from the GitHub repository and may not reflect the latest changes. View repository (github.com, opens in a new tab, external site)
Building a SOTA Exchange for dTAO and Beyond
Powered by TAOS Matrix of Simulations Sandbox
This information was auto-extracted from the GitHub repository and may be incomplete or out of date. Always refer to the subnet's GitHub repository (github.com, opens in a new tab, external site) for the latest instructions.
Sources disagree: Network
Not yet verified: Notes
Not yet verified: VRAM, Notes
Machine-extracted from the subnet's repository docs; may be wrong or out of date. Check the repository before buying or renting hardware. Open the repository (github.com, opens in a new tab, external site) · Extracted 2 Nov 2026
Clone the repository
git clone https://github.com/taos-im/sn-79 && cd sn-79
Execute the miner install script to set up Node.js, pm2, tmux, pyenv, Python 3.10.9, and τaos
./install_miner.sh
Install GenTRX dependencies if participating in distributed training (INSTALL_GENTRX=1 ./install_miner.sh or manual pip install)(optional)
pip install -e . && pip install -e '.[gentrx]'
Create a Bittensor wallet
btcli wallet new_coldkey --wallet.name <coldkey> && btcli wallet new_hotkey --wallet.name <coldkey> --wallet.hotkey <hotkey>
Register the hotkey on subnet 79 (mainnet/finney)
btcli subnet register --netuid 79 --wallet.name <coldkey> --wallet.hotkey <hotkey>
Create R2/Hippius/Storj bucket and commit read credentials on-chain (GenTRX only)(optional)
export GENTRX_AGENT_S3_ACCESS_KEY=<write-key> && export GENTRX_AGENT_S3_SECRET_KEY=<write-secret> && export GENTRX_AGENT_S3_READ_ACCESS_KEY=<read-key> && export GENTRX_AGENT_S3_READ_SECRET_KEY=<read-secret> && python bin/setup_miner_bucket.py --provider r2 --account-id <account-id> --wallet-name <coldkey> --wallet-hotkey <hotkey> --netuid 79 --subtensor-network finney
Set environment variables (S3 credentials for GenTRX, wallet paths, etc.)(optional)
cp .env.example .env && $EDITOR .env
Launch miner (without GenTRX)
./run_miner.sh -w <coldkey> -h <hotkey> -u 79 -a 8091
Launch miner with GenTRX distributed training enabled(optional)
./run_miner.sh -G -w <coldkey> -h <hotkey> -u 79 -a 8091
Clone the repository
git clone https://github.com/taos-im/sn-79 && cd sn-79
Execute the validator install script as root to set up prometheus-node-exporter, Node.js, pm2, tmux, pyenv, Python 3.10.9, vcpkg, g++-14, cmake 3.29.7, and the C++ simulator
./install_validator.sh
Install GenTRX dependencies if participating in distributed training (install CUDA/NVIDIA drivers first for GPU)(optional)
pip install -e . && pip install -e '.[gentrx]'
Create a Bittensor wallet
btcli wallet new_coldkey --wallet.name <coldkey> && btcli wallet new_hotkey --wallet.name <coldkey> --wallet.hotkey <hotkey>
Register the hotkey on subnet 79 (mainnet/finney)
btcli subnet register --netuid 79 --wallet.name <coldkey> --wallet.hotkey <hotkey>
Set S3 bucket credentials as environment variables (GenTRX only)(optional)
export GENTRX_VALIDATOR_S3_WRITE_ACCESS_KEY=<write-key> && export GENTRX_VALIDATOR_S3_WRITE_SECRET_KEY=<write-secret> && export GENTRX_VALIDATOR_S3_READ_ACCESS_KEY=<read-key> && export GENTRX_VALIDATOR_S3_READ_SECRET_KEY=<read-secret>
Configure environment (S3 bucket, API key for gradient server, wallet paths, etc.)(optional)
cp .env.example .env && $EDITOR .env
Launch validator (without GenTRX)
./run_validator.sh -w <coldkey> -h <hotkey> -u 79
Launch validator with GenTRX distributed training enabled (auto-starts local gradient server or connects to remote one)(optional)
./run_validator.sh -G sibling -w <coldkey> -h <hotkey> -u 79
| Day (UTC) | Share |
|---|---|
| 2026-11-04 | 0.02% |
| 2026-11-03 | 0.02% |
| 2026-11-02 | 0.02% |
| 2026-11-01 | 0.03% |
| 2026-10-31 | 0.02% |
| 2026-10-30 | 0.02% |
| 2026-10-29 | 0.02% |
| 2026-10-28 | 0.02% |
| 2026-10-27 | 0.02% |
| 2026-10-26 | 0.02% |
| 2026-10-25 | 0.02% |
| 2026-10-24 | 0.02% |
| 2026-10-23 | 0.02% |
| 2026-10-22 | 0.02% |
| 2026-10-21 | 0.02% |
| 2026-10-20 | 0.02% |
| 2026-10-19 | 0.02% |
| 2026-10-18 | 0.02% |
| 2026-10-17 | 0.02% |
| 2026-10-16 | 0.02% |
| 2026-10-15 | 0.02% |
| 2026-10-14 | 0.03% |
| 2026-10-13 | 0.02% |
| 2026-10-12 | 0.02% |
| 2026-10-11 | 0.02% |
| 2026-10-10 | 0.02% |
| 2026-10-09 | 0.02% |
| 2026-10-08 | 0.02% |
| 2026-10-07 | 0.02% |
| 2026-10-06 | 0.03% |
| 2026-10-05 | 0.02% |
| 2026-10-04 | 0.02% |
| 2026-10-03 | 0.02% |
| 2026-10-02 | 0.02% |
| 2026-10-01 | 0.02% |
| 2026-09-30 | 0.02% |
| 2026-09-29 | 0.02% |
| 2026-09-28 | 0.02% |
| 2026-09-27 | 0.02% |
| 2026-09-26 | 0.03% |
| 2026-09-25 | 0.02% |
| 2026-09-24 | 0.02% |
| 2026-09-23 | 0.02% |
| 2026-09-22 | 0.02% |
| 2026-09-21 | 0.02% |
| 2026-09-20 | 0.03% |
| 2026-09-19 | 0.02% |
| 2026-09-18 | 0.02% |
| 2026-09-17 | 0.03% |
| 2026-09-16 | 0.03% |
| 2026-09-15 | 0.03% |
| 2026-09-14 | 0.03% |
| 2026-09-13 | 0.03% |
| 2026-09-12 | 0.03% |
| 2026-09-11 | 0.03% |
| 2026-09-10 | 0.03% |
| 2026-09-09 | 0.04% |
| 2026-09-08 | 0.04% |
| 2026-09-07 | 0.04% |
| 2026-09-06 | 0.10% |
| 2026-09-05 | 0.28% |
| 2026-09-04 | 0.36% |
| 2026-09-03 | 0.38% |
| 2026-09-02 | 0.34% |
| 2026-09-01 | 0.28% |
| 2026-08-31 | 0.29% |
| 2026-08-30 | 0.29% |
| 2026-08-29 | 0.32% |
| 2026-08-28 | 0.35% |
| 2026-08-27 | 0.38% |
| 2026-08-26 | 0.37% |
| 2026-08-25 | 0.37% |
| 2026-08-24 | 0.39% |
| 2026-08-23 | 0.40% |
| 2026-08-22 | 0.41% |
| 2026-08-21 | 0.40% |
| 2026-08-20 | 0.39% |
| 2026-08-19 | 0.41% |
| 2026-08-18 | 0.44% |
| 2026-08-17 | 0.54% |
| 2026-08-16 | 0.61% |
| 2026-08-15 | 0.70% |
| 2026-08-14 | 0.71% |
| 2026-08-13 | 0.61% |
| 2026-08-12 | 0.61% |
| 2026-08-11 | 0.59% |
| 2026-08-10 | 0.60% |
| 2026-08-09 | 0.62% |
| 2026-08-08 | 0.60% |
| 2026-08-07 | 0.57% |
