DeepSeek-R1 - a chain of thought LLM for the people

It’s great that there is now a competitive, open source chain of thought LLM on par with the Open AI’s model. What I found amazing is that it got the strawberry question correct 😆.

“How many Rs are in ‘strawberry’?”, the model writes:

“First, I’ll spell it out: S-T-R-A-W-B-E-R-R-Y. Now I’ll count: positions 3 (R), 8 (R), and 9 (R). Wait, is that right? Let me check again… Yes, three Rs.”

ChatGPT 3.5 models have often struggled with getting the count right many times. In contrast, ChatGPT 4o successfully solved it on September 12, 2024, and now DeepSeek-R1 achieves the same results—offering accuracy for free and at a fraction of the processing power.

Feel free to read more about it here with an article from Towards AI’s editorial team.