DeepSeek is the talk of the town right now. Even if you don’t use AI like ChatGPT, which you should totally try, you must have heard all the news about the Chinese startup that tanked the stock market. DeepSeek created AI models that are on par with what ChatGPT can do. The DeepSeek R1 reasoning model is what shocked the world.
R1 can match ChatGPT o1, which didn’t have a rival until now. More importantly, DeepSeek R1 training reportedly costs a fraction of the cost of ChatGPT o1.
The Chinese researchers used software optimization to compensate for the lack of hardware. Certain AI-related components, like high-end NVIDIA components, aren’t available to firms like DeepSeek, at least not officially, due to US sanctions.
According to what OpenAI has said recently, DeepSeek might have also used ChatGPT to train early DeepSeek variants without consent. This could explain the fast and affordable development cycle for the R1 precursors.
But let’s not forget that ChatGPT o3 (or o3-mini) is coming soon. That’s the next-gen reasoning model OpenAI unveiled in December, at the end of its “12 Days” AI event. ChatGPT o3’s release date might be imminent, but it’s not (necessarily) related to the arrival of DeepSeek R1.
Instead, the ChatGPT o3 release should happen all according to schedule. OpenAI said in December that the new reasoning model could be out by the end of January after it’s been tested by third-party researchers.
The post Forget DeepSeek AI, when is ChatGPT o3-mini coming out? appeared first on BGR.
Today’s Top Deals
Amazon gift card deals, offers & coupons 2025: Get $500+ free
Today’s deals: $199 iPad 9, Galaxy S25 preorder deals, 25% off COSORI air fryer, $20 Anker power strip, more
Today’s deals: $294 Nintendo Switch OLED, $6 color LED smart bulbs, 26% off Energizer batteries, more
Today’s deals: $329 Apple Watch Series 10, $219 Bose soundbar, 40% off eufy video smart lock, more
Forget DeepSeek AI, when is ChatGPT o3-mini coming out? originally appeared on BGR.com on Thu, 30 Jan 2025 at 06:50:00 EDT. Please see our terms for use of feeds.