OpenAI has published the first measured results of its own Jalapeño… — ITS Crypto - Invest/Trade/Strategy — TG.ME

OpenAI has published the first measured results of its own Jalapeño inference chip

In tests, the accelerator showed 1.5–1.9 times more computing power per watt and 1.7–3.6 times lower latency than systems based on the Nvidia GB200 and GB300.

Subscribe

| YouTube

| TikTok

Exchanges💰: LBank | Blofin | Bitunix | OKX
Get bonuses — sign up on the exchange!