Comment by embedding-shape 5 days ago
Codex, no idea about tokens, I'll upload the session data probably tomorrow so you could see exactly what was done. I pay ~200 EUR/month for the ChatGPT Pro plan, prorating days I guess it'll be ~19 EUR for three days. Model used for everything was gpt-5.2 with reasoning effort set to xhigh.
That 19 EUR figure is basically subscription arbitrage. If you ran that volume through the API with xhigh reasoning the cost would be significantly higher. It doesn't seem scalable for non-interactive agents unless you can stay on the flat-rate consumer plan.
Yeah, no way I'd do this if I paid per token. Next experiment will probably be local-only together with GPT-OSS-120b which according to my own benchmarks seems to still be the strongest local model I can run myself. It'll be even cheaper then (as long as we don't count the money it took to acquire the hardware).
I have my own agent harness, and the inference backend is vLLM.
> I'll upload the session data probably tomorrow so you could see exactly what was done.
I've been very skeptical of the real usefulness of code assistants, much in part from my own experience. They work great for brand new code bases, but struggle with maintenance. Seeing your final result, I'm eager to see the process, specially the iteration.
Thanks in advance, I can't wait to see your prompts and how you architected this...
>I'll upload the session data probably tomorrow so you could see exactly what was done.
That'll be dope. The tokens used (input,output,total) are actually saved within codex's jsonl files.