Ambient Launches Layer 1 Blockchain for Decentralized AI Inference

iconCryptoBriefing
Share
AI summary iconSummary
Ambient has launched a Layer 1 blockchain for decentralized AI inference, using Proof of Logits to verify AI outputs. The network supports large open-weights models and uses an onchain auction system for queries. Built for AI + crypto news, Ambient is compatible with Solana Virtual Machine and targets 65,000 TPS. The 2025-founded project raised $7.2 million in a seed round led by a16z, Delphi Digital, and Amber Group. This blockchain news marks a new approach to AI computation onchain.

Ambient has built a Layer 1 blockchain where the “work” in proof-of-work is replaced by verified AI inference. Instead of hashing, miners process AI queries submitted through an onchain auction system. The network has already served over 134 billion tokens through a public API.

How Proof of Logits actually works

The core innovation is what Ambient calls Proof of Logits, or PoL. When a miner runs an inference task on a large language model, the PoL mechanism verifies that the output was actually generated by the claimed model with the claimed inputs, rather than fabricated or tampered with.

The network currently supports large open-weights models including GLM 5.2 and Kimi K2.7 Code, targeting models with over 600 billion parameters. Queries are processed through a competitive bidding auction, where miners compete on price and speed to fulfill inference requests. The onchain settlement means there’s a verifiable record of every inference task, who processed it, and whether the output checks out.

Advertisement

SVM compatibility and the Solana connection

Ambient is built to be compatible with the Solana Virtual Machine. The chain targets approximately 65,000 transactions per second, borrowing from Solana’s design playbook to handle the high throughput that an AI inference marketplace demands.

Founded in 2025 by Travis Good, the project has moved from concept through testnet to a live environment with a functioning public API.

The case against centralized AI providers

Ambient’s pitch rests on three pillars that centralized AI services struggle with: censorship resistance, privacy, and vendor lock-in. A decentralized inference network distributes the processing across multiple independent miners with no central logging authority. A decentralized marketplace for inference, running open-weights models, eliminates single points of dependency.

What this means for the decentralized AI landscape

One of the more unusual aspects of the project is the reported absence of a native token. In a crypto landscape where tokenomics typically drive speculative interest long before a product ships, building infrastructure without a token suggests Ambient is positioning itself more as an enterprise infrastructure layer, monetizing through API usage rather than token appreciation.

The $7.2 million seed round was raised from a16z’s Crypto Startup Accelerator, Delphi Digital, and Amber Group.

Disclaimer: The information on this page may have been obtained from third parties and does not necessarily reflect the views or opinions of KuCoin. This content is provided for general informational purposes only, without any representation or warranty of any kind, nor shall it be construed as financial or investment advice. KuCoin shall not be liable for any errors or omissions, or for any outcomes resulting from the use of this information. Investments in digital assets can be risky. Please carefully evaluate the risks of a product and your risk tolerance based on your own financial circumstances. For more information, please refer to our Terms of Use and Risk Disclosure.