Inference Labs (INFL)

Preview

Overview

Inference Labs is building cryptographic trust for AI on open networks. The project’s focus is a Proof of Inference protocol that makes AI outputs verifiable with zero-knowledge proofs, so people and apps can rely on model results without revealing model weights or private data. Inference Labs runs Omron, Bittensor Subnet 2 (SN2), and ships tooling that lets developers prove where an AI prediction came from and that it was computed correctly. The goal is simple to state and hard to do: turn AI from a black box into a system you can check with math. (inferencelabs.com)

You’ll sometimes see the phrase “Inference Labs blockchain.” In practice, the team builds on existing blockchains and decentralized networks—primarily Bittensor for AI coordination and Ethereum for smart contracts—rather than launching a separate L1. Their stack sits between AI models and chains, adding proofs, payments, and data integrity. (docs.omron.ai)

Many readers ask about the INFL token and INFL price. As of October 5, 2025, Inference Labs has not publicly published an official “INFL” token launch. On-chain activity around Omron SN2 currently revolves around Ethereum contracts that accept ezETH and WETH deposits and issue a restaking token plus points. If a native INFL token is introduced, its price would be displayed dynamically on this page; no real‑time figures are included here. (docs.omron.ai)

History & Team

Inference Labs Inc. was founded in 2023 in Hamilton, Ontario, by Colin Gagich and Ronald (Ron) Chan. The company announced a $2.3 million pre-seed round on April 15, 2024, led by Digital Asset Capital Management (DACM), Mechanism Capital, and Delphi Ventures, with participation from Big Brain Holdings, Bitscale Capital, Edessa Capital, ID Theory, Metropolis DAO, Mozaik Capital, Newman Capital, and Seed Club Ventures, among others. (globenewswire.com)

In June 2025, the team disclosed that total financing reached $6.3 million, with follow‑on participation from DACM and Delphi Ventures and new investors including Arche Capital, Lvna Capital, and a $1 million Echo Syndicate community round led by Native Capital. The update also highlighted “Proof of Inference” progress and deeper integrations across the decentralized AI stack. (chainwire.org)

Beyond the founders, Inference Labs works with researchers and engineers across cryptography, machine learning, and blockchain. The company publishes research (such as DSperse, a framework for targeted zkML verification) and maintains open tooling like the Proof of Weights SDK for Omron validators. (ar5iv.org)

Notable milestones include:

• Launching Omron on Bittensor (Subnet 2) and deploying complementary Ethereum contracts for deposits and points. (docs.omron.ai)
• Crossing hundreds of millions of zkML proofs verified on-chain by August 2025, signaling scale and ongoing adoption. (inferencelabs.com)

Technology & How It Works

Proof of Inference

Proof of Inference (PoI) is the core idea: AI models produce predictions, and alongside those predictions the system produces a cryptographic proof that the result was generated by a specific model with specific inputs. Inference Labs implements PoI with zero‑knowledge techniques so validators can check integrity without seeing proprietary model weights. This reduces blind trust in AI, similar to how TLS reduced blind trust in websites. (inferencelabs.com)

Omron: Bittensor Subnet 2 and Ethereum

Omron is where the team hardens and scales PoI in production. On Bittensor, two types of nodes participate:

• Miners run small, fast models and return model outputs plus zk proofs.
• Validators query miners, verify proofs, and score responses based on accuracy, proof validity, and performance. (docs.omron.ai)

On Ethereum, SN2 smart contracts accept deposits (currently ezETH from Renzo and WETH) and track point accruals tied to restaking activity. Version 2 expands support by enabling restaking into Renzo directly via SN2 and migrating points from earlier deposits. Together, Bittensor and Ethereum form a loop: Ethereum deposits and restaking provide data and incentives; the Bittensor subnet returns verifiable AI signals for optimizing staking and restaking strategies. (docs.omron.ai)

DSperse and zkML

Full‑model zkML remains expensive. DSperse, introduced by Inference Labs researchers in September 2025, focuses on “targeted verification.” Instead of proving the entire model, DSperse slices neural networks into the layers that matter most and proves only those subcomputations. This keeps proofs practical while preserving confidentiality over key logic. DSperse is modular by design and orchestrates across multiple proving backends. (ar5iv.org)

Developer Tooling and Data Permanence

Developers can use the Proof of Weights SDK to publish validator reward inputs into Omron, linking subnet activity to verifiable artifacts. In addition, Inference Labs has collaborated on “proof publishing” so inference proofs and artifacts can be stored on permanent data networks like Arweave (via AR.IO), preserving provenance of AI results. (github.com)

Tokenomics & Utility

What exists today

As of today, Inference Labs has not announced a public “INFL token” with finalized parameters. Within Omron SN2, users interact with Ethereum smart contracts that support ezETH and WETH deposits. Depositors receive a restaking token and accrue Omron points over time; points reflect participation and feed a broader incentive design for intelligent, verifiable capital networks. In SN2 v2, users can restake through Renzo and continue accruing points as they participate. (docs.omron.ai)

This means any “INFL price” discussion is speculative until an official token is published by the team. For now, activity and rewards are tied to supported assets (ezETH, WETH), a restaking token, and SN2 points within the Omron system. (docs.omron.ai)

Likely roles if an INFL token launches

If Inference Labs introduces an INFL token, “Inference Labs tokenomics” would most likely align with how decentralized AI networks work today. Based on public materials and the system’s architecture, typical roles could include:

• Network utility: paying for proof verification, inference, or data retrieval within the Inference Labs ecosystem.
• Staking and security: bonding by validators/miners to align incentives and prioritize high‑quality, proven outputs.
• Governance: voting on model onboarding standards, proof parameters, or reward curves.
• Incentives: distributing emissions or fees to contributors who improve models, proofs, circuits, and data pipelines.

These functions follow from the project’s own framing of verifiable AI markets and its use of Bittensor plus Ethereum contracts. Final “Inference Labs tokenomics” would depend on an official release. (inferencelabs.com)

Ecosystem & Use Cases

DeFi and restaking

The most mature use case is in DeFi: using verifiable AI to optimize staking and restaking across Actively Validated Services (AVSs). Omron SN2 gathers and analyzes protocol data, then produces cryptographically verified signals that strategy managers can act on, with deposits handled on Ethereum and restaking through partners like Renzo. This is where Inference Labs intersects most directly with “Inference Labs DeFi, NFTs, gaming” keywords: DeFi is live today through SN2’s Ethereum contracts and Bittensor subnet. (docs.omron.ai)

NFTs and media authenticity

For NFTs and digital media, proofs can certify which model created an asset and when, strengthening provenance and reducing silent model swaps. Pairing PoI with permanent storage networks lets creators and marketplaces anchor verifiable metadata directly on-chain and keep it auditable over time. (ar.io)

Gaming and AI agents

In gaming, non‑player characters (NPCs) and AI agents can run with verifiable logic. DSperse’s slice‑and‑prove approach can target only the critical parts of a game agent—like a combat decision layer—making near‑real‑time checks possible without proving an entire large model. This helps studios blend rich AI behavior with on-chain fairness. (ar5iv.org)

Enterprise and public sector

Healthcare, finance, and enterprise AI benefit from confidential inference and audit trails. DSperse emphasizes “verify what matters,” so teams can protect sensitive models while still proving that key subcomputations ran as promised. That aligns with industry demands for privacy, explainability, and compliance‑ready logs. (inferencelabs.com)

Advantages & Challenges

Advantages

• Verifiable AI by default: Proof of Inference replaces trust with math, making AI results checkable on-chain. (cointelegraph.com)
• Dual‑network design: Bittensor provides the competitive market for models; Ethereum provides contracts for deposits, restaking, and points. (docs.omron.ai)
• Targeted zkML: DSperse proves only high‑leverage slices, balancing privacy, performance, and integrity. (ar5iv.org)
• Open tooling: SDKs and docs help validators and builders integrate proof publishing and rewards into their stacks. (github.com)
• Demonstrated scale: Omron has verified hundreds of millions of zkML proofs as of August 2025. (inferencelabs.com)

Challenges

• Complexity: zkML, circuit design, and verifier economics are advanced topics that take time for new builders to adopt at scale. (arxiv.org)
• Evolving standards: Proof systems and hardware acceleration are moving quickly; teams must keep adapting to the latest efficient approaches. (arxiv.org)
• Token clarity: Until an official INFL token is published, third‑party “INFL” listings can create confusion; the project’s primary, verifiable touchpoints remain Bittensor Subnet 2 and Ethereum contracts documented by Inference Labs. (docs.omron.ai)

Where to Buy & Wallets

If you’re researching where to buy INFL, note that Inference Labs has not announced a confirmed, official “INFL token” listing. Today, user interactions with the ecosystem occur through Omron’s Ethereum contracts using ezETH (Renzo) and WETH, with a restaking token and points system. (docs.omron.ai)

When a verified INFL token is published, the common path on EVM networks will look familiar:

• Choose a self‑custody wallet (for example, MetaMask or a hardware wallet) that supports Ethereum and major L2s.
• Fund your wallet with ETH on the correct network to cover swaps and gas.
• Use official Inference Labs links to fetch the correct contract address before any swap.
• Store the asset in a wallet you control; many wallets let you add custom tokens and view balances.

For Omron today, deposits support ezETH and WETH; SN2 v2 adds native restaking into Renzo via the Omron interface. Follow the project’s documentation and interfaces for current steps. (docs.omron.ai)

Regulatory & Compliance

Corporate and network setup

Inference Labs Inc. is a Canadian company based in Hamilton, Ontario. Its products operate across decentralized networks: Bittensor (for AI coordination and incentives) and Ethereum (for contracts). The team has disclosed venture funding rounds in 2024 and 2025 and publishes technical whitepapers and SDKs as part of an open, research‑driven approach. (globenewswire.com)

“Inference Labs halal” and “INFL shariah compliant”

Is Inference Labs crypto halal? Short answer: No—there is no published halal certification or shariah review for Inference Labs or an “INFL token” on official channels as of October 5, 2025. If a token is formally launched later, its shariah status would depend on the token’s design, revenue sources, and how funds are used; at present, no certification has been announced. (inferencelabs.com)

Inference Labs regulatory status

There has been no public claim of special licensing for token issuance. The project integrates with public networks (Bittensor, Ethereum) and partners (for example, Renzo for restaking) while focusing on cryptographic verification and infrastructure. Any eventual INFL token—if launched—would need to reflect applicable rules in the jurisdictions where it is offered and traded; as of now, the company’s public materials center on infrastructure, not on a regulated token sale. (docs.omron.ai)

Future Outlook

Inference Labs’ roadmap points to deeper protocol integrations and broader coverage of AI models with practical proofs. On SN2, the team is expanding model support, shifting to output‑based scoring, and wiring Ethereum contracts more tightly into restaking flows, including Renzo. The broader Proof of Inference effort is being positioned to interoperate across networks, so proofs and verified outputs can flow where they add the most value. (docs.omron.ai)

On the research side, DSperse signals a pragmatic path for zkML by verifying only the most important slices of an inference pipeline. This approach fits real‑world needs across DeFi, NFTs, gaming, and enterprise AI. We also see early moves to make proof artifacts permanent and queryable, so the “receipt of intelligence” lives on-chain and on permanent data layers. If a native INFL token appears, expect it to reinforce these loops: fund proofs, reward accurate models, and govern ever‑larger networks of verified intelligence. (ar5iv.org)

Summary

Inference Labs is building a trust layer for AI with cryptographic proofs, not promises. The team runs Omron, Bittensor’s Subnet 2, and deploys Ethereum contracts that tie restaking activity to verifiable AI signals. Research like DSperse makes zkML more practical by proving only what matters. While an official INFL token has not been announced yet, the foundations for “Inference Labs tokenomics” are visible in how SN2 coordinates miners, validators, deposits, and proofs. As adoption grows across DeFi and beyond—from NFTs to gaming and enterprise—the value of verifiable, tamper‑evident AI will likely define the next wave of on‑chain intelligence. Watch this page for any future updates on INFL price, where to buy INFL, and the evolving Inference Labs blockchain ecosystem as it matures. (docs.omron.ai)