What is Dolphin (POD)?

By CMC AI
02 October 2026 09:32PM (UTC+0)
TLDR

Dolphin (POD) is a cryptocurrency project building a decentralized AI inference network and developing open-source AI models, with its POD token facilitating staking and bonding within the ecosystem.

  1. It operates a distributed GPU pool to provide scalable and private AI inference services.

  2. The team develops and releases uncensored, instruct-tuned large language models like Dolphin 3.0.

  3. The POD token is used for staking to secure the network and bonding to access compute resources.

Deep Dive

1. Purpose & Value Proposition

Dolphin addresses the limitations of centralized AI services like ChatGPT by giving users full control. According to the project's documentation, centralized providers control system prompts, model versions, alignment, and user data, which can break software and compromise privacy (dphn/Dolphin3.0-Qwen2.5-3b). In contrast, Dolphin's decentralized network lets users set their own guidelines and keep data private while accessing distributed GPU power for AI tasks.

2. Technology & Architecture

The project has two core technical pillars. First, it develops open-source AI models; the Dolphin series are instruct-tuned models capable of coding, math, and agentic tasks, released on platforms like Hugging Face. Second, it operates a distributed inference network that assigns user requests to available GPU nodes across a pool, optimizing resource use (CoinMarketCap). The network uses the Base blockchain for token transfers.

3. Tokenomics & Utility

POD has a total supply of 500 million tokens (Poloniex). Its primary utilities are staking—likely to secure the network and reward node operators—and bonding, which may be required to access inference services or participate in governance. This aligns the token with the network's operational security and demand.

Conclusion

Dolphin is fundamentally a decentralized AI infrastructure project that merges open model development with a distributed compute network. Can its dual focus on creating state-of-the-art models and maintaining a robust inference network drive sustainable adoption?

CMC AI can make mistakes. Not financial advice.