ChatGPT sub comparison - work mode

[Superjakes]

I got into 3D printing recently and at the same time, I took up a free month of ChatGPT Plus, which turned out super useful in managing the learning curve of my new hobby. After the month was over, I decided to keep a subscription, but downgrade to the R150 per month ChatGPT Go. Turns out the work mode allowance is really subpar in there, however the chat mode is almost bottomless. So for diagnosing visual problems, it is ok, but for creating stuff (any kind of documentation or design), it is not feasible.

Now for the question part - does anybody know if the Plus option would truly be better in terms of work mode tokens? It does not seem to be well documented anywhere. The Go work mode tokens can comfortably get burnt through in 20 minutes, with no proper resolve of the task posed to it. I don't want to throw more money at it if I am not sure it will solve my problem.

 

[Kosmik]

Each sub burns token as a base. The chat is normally much higher and will drop you to a lower model accordingly. Work mode, is very similar if not the same as codex, its running on your machine or a virtual and will consume tokens via calls in some metered method. The exact metering is not disclosed but you can comfortably say that the pool is increased based on your tier : go, plus ,pro etc.

Unfortunately how well it will do for you, that's tricky. I can understand the need, I had mine do a design stl for me for a person needing a toy and it did it quite well and produced artifacts etc. I didn't do it via codex/work though. In the end, for 3d its just a stl. If you want it to do more than that or work local mcp with some design, then its going to cost tokens.

 

[saor]

Can you be a bit more specific about what you're using it for? I find Gemini cheaper for some things, GPT Terra cheaper for some things, Deepseek cheaper for most things etc

 

[Superjakes]

Can you be a bit more specific about what you're using it for? I find Gemini cheaper for some things, GPT Terra cheaper for some things, Deepseek cheaper for most things etc

I have some very niche educational toy prints. Developed them during my free Plus month, and adjusted them with different content as per customer request. Basically changing the lettering, that type of thing. So input - STL/3MF file, add prompt and get adjusted file back. If it is only one iteration, I'd be more ok with it, but more often than not, I need to take it back and ask it to rather do it the way I asked, etc...