gpt-5.6-terra vs sonar-pro
gpt-5.6-terra is 1.5× cheaper on input than sonar-pro.
| gpt-5.6-terra | sonar-pro | |
|---|---|---|
| Input / 1M | $2.00 | $3.00 |
| Cached input / 1M | $0.20 | — |
| Output / 1M | $12.00 | $15.00 |
| Context window | 1.1M | 200K |
| Max output | 128K | 8K |
| Provider | OpenAI | Perplexity |
| Vision | Yes | No |
| Tool calling | Yes | No |
| Token count | exact | estimated |
Capability for the money
Independent measurement · Artificial Analysis · read 2026-08-11gpt-5.6-terra is 1.3× cheaper than sonar-pro and scores higher on hard reasoning (GPQA Diamond) — there is no case for paying more here.
Hard reasoning
graduate-level science and multi-step analysis
GPQA Diamondgpt-5.6-terra +34.7
Averaged over the benchmarks both models sat, gpt-5.6-terra returns 20.6 points of pass rate per dollar of blended price against 9.6 — the ratio that decides whether the dearer model earns its rate.
Each figure is a pass rate: the share of that test set the model answered correctly, measured by a third party rather than self-reported by the lab. Rows are different tests of different difficulty, so compare a model against the other model, not one row against the next. No published following instructions, coding or tool use score for one of these models, so those lenses are omitted. And a benchmark is not your workload — it narrows the shortlist, it does not pick for you.
Which to pick
- Cheaper input: gpt-5.6-terra
- Cheaper output: gpt-5.6-terra
- Larger context: gpt-5.6-terra
- Better at hard reasoning: gpt-5.6-terra (92.5% vs 57.8% on GPQA Diamond)
List price is not production cost. Output length, cache hit rate, retries and task quality move the real number more than the headline rate does — price your own prompt on the token ledger before deciding.
Related comparisons
gpt-5.6-terra detail · sonar-pro detail · All OpenAI pricing · All comparisons