.disclaimer I’m not a developer, but I have some familiarity with Python and language models. The concept of creating a locally fine-tuned Language Model from either llama2 or mistral named Electos (chosen) for Solana, designed to work exclusively with Solana smart contracts and wallets, seems intriguing. It could potentially be powerful, especially for applications like gaming, NFT role-playing, or renting out larger models for collaborative communication. The way I kinda envisioned The process involves downloading the fine-tuned model onto a wallet, running it locally using your CPU or GPU (similar to LLM Studio), with personal use being free, such as for NFT character roleplay. However, there might be a fee for outsourcing, or having multiple llm communicate depending on the computational strength required. I don’t know if this idea is feasible; perhaps others can share their opinions. I seriously considered tackling it myself, but given my lack of knowledge with Rust and being a self-taught coder with average skills probably take maybe a while. Would love to hear different perspectives