Currently I would not invest in any solution that enforces usage of a single LLM model provider, as the scene is really dynamic with the best player changing every month. Now Gemini 2.5 looks really great, but Anthropic, OpenAI, DeepSeek and others are probably working on the next, better model.
So, Aider + whatever seems to be working best, seems to be the best strategy that works for me. Aider has this nice feature that I can specify what files are added to the context, so I pay less and I do not confuse AI with stuff unrelated to a given task.
So, Aider + whatever seems to be working best, seems to be the best strategy that works for me. Aider has this nice feature that I can specify what files are added to the context, so I pay less and I do not confuse AI with stuff unrelated to a given task.