GPT-4o vs GPT-5.6 Terra: cost compared
Last verified
GPT-5.6 Terra is 20% cheaper on input and -20% cheaper on output than GPT-4o. Whether that makes it the right choice depends entirely on whether it does your task well enough — this page gives you the cost side of that trade, precisely.
Rates side by side
| $ per million tokens | GPT-4o | GPT-5.6 Terra |
|---|---|---|
| Input | $2.50 | $2.00 |
| Output | $10.00 | $12.00 |
| Cached input | $1.25 | $0.20 |
| Output : input ratio | 4.0× | 6.0× |
What the difference is worth
Monthly cost for the same workload on each, without caching:
| Workload | GPT-4o | GPT-5.6 Terra | Difference |
|---|---|---|---|
| Support chatbot 1,000 conversations/month · 2,000 in + 500 out each | $10.00 | $10.00 | $0.0000 saved |
| RAG search 10,000 queries/month · 8,000 in + 400 out each | $240 | $208 | $32.00 saved |
| Bulk extraction 10,000 documents · 3,000 in + 300 out each | $105 | $96.00 | $9.00 saved |
| Coding agent 100 runs/month · 400,000 in + 20,000 out each | $120 | $104 | $16.00 saved |
Which to use
GPT-5.6 Terra if the task is well-specified and mechanical — extraction, classification, formatting, routine transformation. The saving is real and compounds with volume.
GPT-4o if the task involves genuine reasoning, long-horizon planning, or work where a wrong answer is expensive to catch. Paying 1.3× more on input is trivial compared to the cost of shipping a bad result.
Both are OpenAI models, so switching is usually a one-line change and the tokenizer is the same — which makes this a genuinely easy experiment to run. Test on your own workload before deciding.
Don't skip caching
Cached input costs $1.25 on GPT-4o and $0.20 on GPT-5.6 Terra. If your prompts share a stable prefix, caching on the more expensive model can beat switching to the cheaper one outright — worth checking before you migrate anything.
Sources
OpenAI rates verified 2026-08-05 (source). OpenAI rates verified 2026-08-05 (source).