OpenAI has developed a new chip, Jalapeño, designed for fast and scalable artificial intelligence inference. Benchmarks demonstrate its superior performance compared to current state-of-the-art technology.
According to reporting from TechCrunch, the Jalapeño chip was tested on SemiAnalysis’ InferenceX benchmark. The results showed that Jalapeño registered both more tokens per user and more throughput per kilowatt than existing leading chips.
The testing indicates that the chip is optimized for efficiently processing AI requests at a large scale. The benchmarks suggest that Jalapeño could be a significant advancement in the field of AI hardware, enabling faster and more cost-effective AI deployments.
Read the original coverage
💬 Comments
📜 Comment Policy