Model Surprising Users with 2025-Level Reasoning
DeepSeek is a really good model, but it is not necessarily better than GPT-4o or Claude overall. However, since it is both free and gaining a lot of attention, many users who previously relied on free "mini" models are now experiencing what an early 2025 reasoning AI can do—and they are surprised by its capabilities.
Additionally, I have been testing it using [some testing tools] and have observed significant efficiency and strong LLM outputs. One of my friends, Mr. Gary Marcos, who works at Qualcomm, has also shared similar insights.
Talking product sense with Ridhi
9 min AI interview5 questions
The important thing is that it's open-source so someone is going to jailbreak and then put the uncensored model out on the web.
If you're serious about leveling up your DSA & System Design skills and aiming for top product-based companies, this is your golden opportunity! JOIN NOW👉🏽 http://bit.ly/3WGvizg