OpenAI's Jalapeo, its first custom AI inference chip, delivers up to 1.9 more work per watt and 3.6 lower latency. The chip is set for deployment by late 2026, with next generations already in development.
Market Samachar is a news aggregator. This article was originally published by NDTV Profit - Latest. Tap the button above to read the full story on their site.