Verdict
Built from the measurements below
- Accuracy is too close to call: GPT-6.1 Terra scored 73.8% and o3 72.5% on the same 80 questions, a gap within what chance alone could produce.
- GPT-6.1 Terra is 6.9× cheaper per question as billed ($0.0013 vs $0.0088), reasoning tokens included.
- o3 generates text 1.3× faster: 163 tokens per second against 122.
- o3 starts answering sooner: 1.2 s to the first token, against 2.4 s for GPT-6.1 Terra.
- GPT-6.1 Terra accepts more context: 1.1M tokens against 200k.
Which should you choose?
Choose GPT-6.1 Terra for the lower cost per correct answer ($0.0017 against $0.012), shorter outputs (about 195 tokens per question against 1,025) and a 1.1M-token context window. Choose o3 for faster streaming and a quicker first token. The widest subject gap is in law: o3 answered 5 of 6 correctly, GPT-6.1 Terra 3 of 6. With so few questions per subject, treat this as a lead to test on your own prompts, not a result.
Accuracy gap +1.3 points (z = 0.18). We call a winner only when z is at least 1.96, the usual 95% threshold. Speed, latency and price need a 10% margin.