tngtech

The German TNG has thrown the new R1 into the model soup (because DeepSeek released an updated R1 model on May 28th), improving performance over the original Chimera model even further.

A model merge of DeepSeek R1 and DeepSeek V3. It works surprisingly well, resulting in a model that has shorter CoTs and thus is faster to run, while still being more capable than just using V3. Even more surprising: The merged model refuses fewer requests than both V3 and R1.