[Vitalik Updates on CROPS AI Progress: Proposes Combining Private Remote LLM Calls with Ethereum Private RPC Reads] Deep Tide TechFlow reports that on May 28, Vitalik Buterin shared multiple updates on CROPS AI developments, highlighting a significant overlap between the 'CROPS Ethereum Access Layer' and 'CROPS AI.' He stated that if remote large language models (LLMs) could be accessed via paid calls using zero-knowledge proofs, it could also address the issue of Ethereum private RPC reads. Meanwhile, Deepseek v4 has been released, with a 2-bit quantized version capable of running within 90 GB, though it currently performs faster on Apple hardware. The messaging app messa now supports Telegram Alpha features. The model execution tool luceb has demonstrated higher efficiency when running computation-intensive models. The local AI recording tool VoxTerm is still under active development. Vitalik Buterin also mentioned that fine-tuning large language models for specific applications could enhance secure code-writing capabilities, and Ethereum-related scenarios should have corresponding fine-tuned models.