Its a great update, I ran a benchmark comparing verbosity and intelligence compared to old weights.
- Reasoning effort = 1 is now actually usable and produces 2.5x less tokens than old weights, without intelligence lost.
- Overall 20% less tokens at each reasoning effort.
- Same intelligence and same tool-calling pattern.
More details in blog post https://sference.com/blog/deepseek-v41f-1017-benchmark
Improved on exactly the items that I hear people complain about with the previous checkpoint!