Subnet Directory

With the amount of new subnets being added it can be hard to get up to date information across all subnets, so data may be slightly out of date from time to time

Subnet 109

Taoillium

Emissions
Value
Recycled
Value
Recycled (24h)
Value
Registration Cost
Value
Active Validators
Value
Active Miners
Value
Active Dual Miners/Validators
Value

ABOUT

What exactly does it do?

Taoillium (Subnet 109) is a decentralized, blockchain-integrated AI agent marketplace designed to unify the creation, deployment, and reward mechanisms for intelligent software agents. In broad terms, Taoillium allows developers to build autonomous agents—services like chatbots, recommendation engines, or arbitrage bots—such that these agents can be monetized through Bittensor’s TAO token economy and made interoperable within the growing AI ecosystem.

Autonomous agent publishing: Developers submit AI agents to the Taoillium registry. Each agent is identified by a unique UID and encapsulates its logic—this might be a Python-coded agent, a Transformer-based conversational bot, or a domain-specific decision-making model. Agents are packaged in containers (via Slack integration like Docker), ensuring portable and standardized execution.

Invocation via wallet spending: Bittensor wallets can invoke agents by spending on their UID. This translates to a transaction on Subnet 109, where TAO tokens are staked and transferred to the agent. The act of spending TAO initiates the agent’s execution on an off-chain compute node (hosted by the developer or Taoillium infrastructure), which performs the designated task—responding to a prompt, running logic, or performing an API operation.

Reward and feedback loop: After execution, the agent returns a result (e.g., a text response, calculation, or API output) tied to the invocation ID. The total TAO spent is awarded to the agent’s UID and stored in the subnet’s treasury as metadata. This creates a performance-incentivized economy, where useful agents accumulate TAO, encouraging high-quality and responsive behaviors from developers.

Embrace of Web3-native applications: Taoillium integrates seamlessly with Bittensor wallets and UI frameworks, allowing users to explore, interact with, and invest in the agents they find valuable. The UI visualizes active agents, their reputations (based on TAO accrued), and invocation stats, mimicking a decentralized app store powered by blockchain and token incentives.

Overall, Taoillium transforms the process of building and monetizing AI agents into a trustless, permissionless, and interoperable environment. Developers maintain full custody of code and hosting, while users discover agents on-chain, pay per use, and reward agent creators via TAO tokens. This ecosystem addresses challenges in distribution, monetization, and governance for AI services within decentralized AI networks.

 

Taoillium (Subnet 109) is a decentralized, blockchain-integrated AI agent marketplace designed to unify the creation, deployment, and reward mechanisms for intelligent software agents. In broad terms, Taoillium allows developers to build autonomous agents—services like chatbots, recommendation engines, or arbitrage bots—such that these agents can be monetized through Bittensor’s TAO token economy and made interoperable within the growing AI ecosystem.

Autonomous agent publishing: Developers submit AI agents to the Taoillium registry. Each agent is identified by a unique UID and encapsulates its logic—this might be a Python-coded agent, a Transformer-based conversational bot, or a domain-specific decision-making model. Agents are packaged in containers (via Slack integration like Docker), ensuring portable and standardized execution.

Invocation via wallet spending: Bittensor wallets can invoke agents by spending on their UID. This translates to a transaction on Subnet 109, where TAO tokens are staked and transferred to the agent. The act of spending TAO initiates the agent’s execution on an off-chain compute node (hosted by the developer or Taoillium infrastructure), which performs the designated task—responding to a prompt, running logic, or performing an API operation.

Reward and feedback loop: After execution, the agent returns a result (e.g., a text response, calculation, or API output) tied to the invocation ID. The total TAO spent is awarded to the agent’s UID and stored in the subnet’s treasury as metadata. This creates a performance-incentivized economy, where useful agents accumulate TAO, encouraging high-quality and responsive behaviors from developers.

Embrace of Web3-native applications: Taoillium integrates seamlessly with Bittensor wallets and UI frameworks, allowing users to explore, interact with, and invest in the agents they find valuable. The UI visualizes active agents, their reputations (based on TAO accrued), and invocation stats, mimicking a decentralized app store powered by blockchain and token incentives.

Overall, Taoillium transforms the process of building and monetizing AI agents into a trustless, permissionless, and interoperable environment. Developers maintain full custody of code and hosting, while users discover agents on-chain, pay per use, and reward agent creators via TAO tokens. This ecosystem addresses challenges in distribution, monetization, and governance for AI services within decentralized AI networks.

 

PURPOSE

What exactly is the 'product/build'?

Taoillium is primarily a protocol layer atop Bittensor, supported by a solid developer toolkit and ecosystem infrastructure. Its product comprises five key layers:

Subnet‑109 Protocol & Registry

  • Built using Bittensor’s Subtensor framework, the on-chain registry stores agent metadata: UIDs, deployment endpoints, spendable status, and developer public keys.
  • Invoke transactions are Bittensor operations that transfer TAO from wallet hotkeys to agent UIDs, triggering off-chain execution while securely logging on-chain transfers.
  • The mechanism supports batch invocation and enforces essential logic (e.g., entry price, daily spend caps, and trust thresholds) via consensus code.

 

Agent Connectors & Wrappers

  • Taoillium offers lightweight Python/Node.js agent wrappers (on GitHub/SDK) that conform to a standardized schema—methods like on_invoke(invocationID, payload) receive inbound invocations and return appropriately structured responses.
  • These wrappers handle crucial details including payload decryption, response packaging, billing integration, and webhook or HTTP endpoint triggers.

 

Off‑Chain Executors / Agent Runtime

  • A key component is the runtime host (often container-based) which listens to Subnet 109 for new invocation events.
  • On invocation, it triggers agent execution (could be calling local ML models, external APIs, or user-defined computation).
  • After execution, it constructs a response transaction containing results (text, JSON data, etc.) along with the DAO reference, which is sent back on-chain and then relayed to the invoking wallets.

 

Developer Dashboard & CLI Tools

  • The Taoillium app dashboard (hosted at app.taoillium.ai) enables developers to register agents, set pricing, monitor invokes, and manage agent state (active, paused, deprecated).
  • The CLI tools allow developers to: Register UIDs for their agents, set traffic limits (daily cap on TAO spends) and view earnings and invocation logs
  • Public documentation contains code samples showing how to integrate agent handlers and respond to invocation webhooks.

 

Economy & Token Flow

  • Each invocation leads to a micro-payment of TAO into the agent’s wallet, recorded in Subnet 109 balances.
  • TAO remains in the sender’s wallet until the agent responds. On successful execution, the tokens are transferred to the agent’s developer wallet as a direct reward.
  • If an agent fails or times out, TAO might be returned or burned, depending on subnet rules. These financial flows are visible and auditable on-chain.

 

Technically, the system relies on an event-listener architecture: wallet invocations raise events, agents react off-chain, and response transactions finalize the call-and-response cycle. This implements a true serverless AI function framework but operates fully within the blockchain’s economic and auditing infrastructure. The code leverages async, event-driven Python (based on frameworks like FastAPI), container orchestration tools (Docker), and Bittensor RPC interfaces.

 

Taoillium is primarily a protocol layer atop Bittensor, supported by a solid developer toolkit and ecosystem infrastructure. Its product comprises five key layers:

Subnet‑109 Protocol & Registry

  • Built using Bittensor’s Subtensor framework, the on-chain registry stores agent metadata: UIDs, deployment endpoints, spendable status, and developer public keys.
  • Invoke transactions are Bittensor operations that transfer TAO from wallet hotkeys to agent UIDs, triggering off-chain execution while securely logging on-chain transfers.
  • The mechanism supports batch invocation and enforces essential logic (e.g., entry price, daily spend caps, and trust thresholds) via consensus code.

 

Agent Connectors & Wrappers

  • Taoillium offers lightweight Python/Node.js agent wrappers (on GitHub/SDK) that conform to a standardized schema—methods like on_invoke(invocationID, payload) receive inbound invocations and return appropriately structured responses.
  • These wrappers handle crucial details including payload decryption, response packaging, billing integration, and webhook or HTTP endpoint triggers.

 

Off‑Chain Executors / Agent Runtime

  • A key component is the runtime host (often container-based) which listens to Subnet 109 for new invocation events.
  • On invocation, it triggers agent execution (could be calling local ML models, external APIs, or user-defined computation).
  • After execution, it constructs a response transaction containing results (text, JSON data, etc.) along with the DAO reference, which is sent back on-chain and then relayed to the invoking wallets.

 

Developer Dashboard & CLI Tools

  • The Taoillium app dashboard (hosted at app.taoillium.ai) enables developers to register agents, set pricing, monitor invokes, and manage agent state (active, paused, deprecated).
  • The CLI tools allow developers to: Register UIDs for their agents, set traffic limits (daily cap on TAO spends) and view earnings and invocation logs
  • Public documentation contains code samples showing how to integrate agent handlers and respond to invocation webhooks.

 

Economy & Token Flow

  • Each invocation leads to a micro-payment of TAO into the agent’s wallet, recorded in Subnet 109 balances.
  • TAO remains in the sender’s wallet until the agent responds. On successful execution, the tokens are transferred to the agent’s developer wallet as a direct reward.
  • If an agent fails or times out, TAO might be returned or burned, depending on subnet rules. These financial flows are visible and auditable on-chain.

 

Technically, the system relies on an event-listener architecture: wallet invocations raise events, agents react off-chain, and response transactions finalize the call-and-response cycle. This implements a true serverless AI function framework but operates fully within the blockchain’s economic and auditing infrastructure. The code leverages async, event-driven Python (based on frameworks like FastAPI), container orchestration tools (Docker), and Bittensor RPC interfaces.

 

WHO

Team Info

Awaiting content…

Awaiting content…

FUTURE

Roadmap

Alpha Launch (Q2 2025)

  • Core functionality delivered: agent registration, invocation API, basic app interface (dashboard).
  • Developers invited to build, register, and test agents live using test TAO.

 

Beta Growth (Q3 2025)

  • Expand SDK coverage (Node.js, Rust).
  • Introduce invoicing templates for better payload extensibility.
  • Add advanced monetization tools: sliding pricing, subscription models, usage tiers, and reputation signals such as graded reviews or ratings.
  • Implement batching and gas optimization, enabling micro-invokes with shared on-chain funds (e.g., invoking an agent multiple times in a single transaction).

 

Inter-Subnet Integration (Q4 2025)

  • Allow agents on other subnets (e.g., Subnet 12 or 63) to internally call Taoillium agents and potentially re-sell their services (e.g., a quantum agent using a language API).
  • Introduce cross-subnet call support such that agents can pass payloads and TAO across subnet boundaries for on-demand microservice invocation.
  • Marketplace & Discoverability (2026)
  • Release an open marketplace interface: search, filter, and evaluate agents.
  • Introduce agent reputation scoring and on-chain rating system.Onboard select third-party agents (weather stations, analytics services, ML inference models) as examples of advanced capabilities.

 

Governance & DAO Integration (H2 2026)

Launch Taoillium DAO to govern:

  • New agent approval
  • Revenue-sharing policies
  • Dispute resolution
  • Token governance: potential creation of a META-like agent token with shared ownership over marketplaces revenue, and treasury pooling for hackathons and educational grants.

 

Enterprise & Compliance (2027+)

  • GDPR compliance and KYC/AML integrations for sensitive use cases.
  • Marketplace adoption by enterpriseing blockchain native apps for business services and offers.
  • API partnerships with existing AI platforms, IoT fleets, and automation tools to embed Taoillium agents.

 

Beyond these, Taoillium has proposed support for multi-model agents (e.g. chaining multiple AI models in one agent flow) and plug-in infrastructure where external services like Pinecone, FastAPI, or cloud endpoints can be built into a serverless agent. However, these advanced builds are tagged “2026+” in their strategic notes.

 

Alpha Launch (Q2 2025)

  • Core functionality delivered: agent registration, invocation API, basic app interface (dashboard).
  • Developers invited to build, register, and test agents live using test TAO.

 

Beta Growth (Q3 2025)

  • Expand SDK coverage (Node.js, Rust).
  • Introduce invoicing templates for better payload extensibility.
  • Add advanced monetization tools: sliding pricing, subscription models, usage tiers, and reputation signals such as graded reviews or ratings.
  • Implement batching and gas optimization, enabling micro-invokes with shared on-chain funds (e.g., invoking an agent multiple times in a single transaction).

 

Inter-Subnet Integration (Q4 2025)

  • Allow agents on other subnets (e.g., Subnet 12 or 63) to internally call Taoillium agents and potentially re-sell their services (e.g., a quantum agent using a language API).
  • Introduce cross-subnet call support such that agents can pass payloads and TAO across subnet boundaries for on-demand microservice invocation.
  • Marketplace & Discoverability (2026)
  • Release an open marketplace interface: search, filter, and evaluate agents.
  • Introduce agent reputation scoring and on-chain rating system.Onboard select third-party agents (weather stations, analytics services, ML inference models) as examples of advanced capabilities.

 

Governance & DAO Integration (H2 2026)

Launch Taoillium DAO to govern:

  • New agent approval
  • Revenue-sharing policies
  • Dispute resolution
  • Token governance: potential creation of a META-like agent token with shared ownership over marketplaces revenue, and treasury pooling for hackathons and educational grants.

 

Enterprise & Compliance (2027+)

  • GDPR compliance and KYC/AML integrations for sensitive use cases.
  • Marketplace adoption by enterpriseing blockchain native apps for business services and offers.
  • API partnerships with existing AI platforms, IoT fleets, and automation tools to embed Taoillium agents.

 

Beyond these, Taoillium has proposed support for multi-model agents (e.g. chaining multiple AI models in one agent flow) and plug-in infrastructure where external services like Pinecone, FastAPI, or cloud endpoints can be built into a serverless agent. However, these advanced builds are tagged “2026+” in their strategic notes.

 

NEWS

Announcements

Taoillium Follow 239 658

AI subnet on #Bittensor for high-speed | high-availability | performance-verified inference

Taoillium
Load More

View other subnets

Back to the Directory