I agree, from my experience Grok gives superior coding results, especially when modifying large sections of the codebase at once such as in refactoring.
Although it’s not for coding, I have noticed Gemini 2.5 pro Deep Research has surpassed Grok’s DeepSearch in thoroughness and research quality however.
Yes, Grok has become my go to model for general research and targeted coding tasks. Feels like its getting better over time vs ChatGPT which seemed to deteriorate over time.
Claude 3.7 is excellent, and better at coding but I appreciate the context size of Grok and feel like I get better bang for my buck for general purpose research too.
I don’t know if I’m doing something wrong, but Gemini 2.5 pro was substantially worse coding quality than Grok. Which is surprising since I’m working on a Golang codebase which I had assumed Gemini would excel at given that it’s made by Google
The trick I’ve been using is to copy the entire codebase into a text prompt with Repo Prompt and feed that into Grok with a specific request on what feature / change I want.
Then paste that output into Cursor with Claude 3.7 and have it make the actual code changes and ask it to build/fix errors along the way with yolo mode enabled.
The 2-step process is a lot better since Grok can refer to the entire context of your codebase in one shot and come up with a high quality implementation plan, which is then handed off to Cursor to autonomously make the code changes.
I believe it is fairly accurate at estimating the ride time. The first time I took it though, I mistakenly thought it would take the same amount of time to get to my destination as Google Maps estimated since that is usually the case for Uber/Lyft.
So I was pretty annoyed after I got into the car and then realized that the route it selected was going through a bunch of hills and side streets that would take twice as long as the most direct route (via Google Maps) and there's nothing I could do to change that once the ride started.
I could definitely see that. If you're not pressed on time it can be a smoother ride for sure. But if you're trying to get somewhere fast or efficiently like going to the airport, you'd want to get a regular Lyft/Uber otherwise you might miss your flight.
It would be cool if you can configure the ride preferences for how aggressive you'd like it to drive.
In my experience Waymo has been much worse compared to Lyft/Uber for longer rides.
A big problem with Waymo in its current state is how its routes are terribly inefficient.
It purposely avoids freeways and higher speed roads, opting to take more inefficient routes without regards to the number of stop signs, hills, and other factors which will inevitably lead to a longer travel time. It's almost like it's using a worse version of the "Avoid highways" feature on Google Maps, and getting to a further destination can take almost twice the amount of time as compared to a Lyft/Uber.
Another problem is its lack of human intuition and strategies when driving in the city during some kind of event where many of the roads are blocked off. A human driver would have been navigating the blocked roads throughout the day and already know where to go to avoid the crowds, where as Waymo naively follows its navigation system and gets stuck in a bunch of traffic for no reason.
It also drives annoyingly slowly which leads to frustration from human drivers who constantly try to overtake you.
Although it’s not for coding, I have noticed Gemini 2.5 pro Deep Research has surpassed Grok’s DeepSearch in thoroughness and research quality however.