TechQwen 3.8 Flash reportedly runs at up to 373 tokens per second on a laptop
CryptoSlate reports that Vitalik Buterin sees local AI handling more work without sending users’ full personal context to the cloud—but argues crypto wallets still need deterministic transaction limits and human approval.
TLDR
CryptoSlate reports that Vitalik Buterin’s latest tests showed Qwen 3.8 Flash running at up to 373 tokens per second on a laptop. He says local AI is getting fast enough to handle much more work without sending full personal context to the cloud. For crypto wallets, though, he argues speed and privacy are not enough: transaction authority still needs deterministic limits and human approval.
Combined views
1.7K
1 Source, first seen 1d ago
Qwen 3.8 Flash reportedly runs at up to 373 tokens per second on a laptop
CryptoSlate reports that Vitalik Buterin sees local AI handling more work without sending users’ full personal context to the cloud—but argues crypto wallets still need deterministic transaction limits and human approval.
TLDR
CryptoSlate reports that Vitalik Buterin’s latest tests showed Qwen 3.8 Flash running at up to 373 tokens per second on a laptop. He says local AI is getting fast enough to handle much more work without sending full personal context to the cloud. For crypto wallets, though, he argues speed and privacy are not enough: transaction authority still needs deterministic limits and human approval.