Here is some high level comparison between these two in LM Studio.
deepseek-r1-distill-qwen-7b
Session Title: Learning Path to Programming with Python Focus
Thought for 33.15 seconds
deepseek-r1-distill-llama-8b
Session Title: Programming Fundamentals
Thought for 20.74 seconds
With the extra time for thinking, the qwen version settled on a Python/Flask focus for the learning and provided specific code examples. That focus is then reflected in the session title. The llama version just provides general fundamentals with no code samples at all, though they were specifically requested. I’m not sure if that difference is due to the extra time or just the different family.