GPT-6.1 Sol
Previous generalist. September’s everyday model, still sold at the same price as GPT-6.2 Sol.
Which reasoning effort setting?
controls how long GPT-6.1 Sol thinks before answering. Token prices never change; higher settings spend more and raise task cost.
Start at Max. The top setting is also the best value here. Rule: cheapest setting within 2 points of the model’s best score.
At Max, GPT-6.1 Sol thinks for about 49.2K tokens per task, then writes about 18.6K more as its answer: $1.27 per task, 7.9× Low, for +19.8 points.
Output rate: $10.00/1M at every effort level. Thinking tokens are billed as output.
| Effort | Index | Per task | Thinking | All output | × Low | Time |
|---|---|---|---|---|---|---|
| Low | 34.8 | $0.16 | 1.6K | 5.8K | 1.00× | — |
| Medium | 46.3 | $0.36 | 4.8K | 13.5K | 2.23× | — |
| High | 50.3 | $0.59 | 10.1K | 22.6K | 3.67× | — |
| Xhigh | 52.5 | $0.76 | 15.5K | 30.2K | 4.75× | — |
| MaxStart here | 54.6 | $1.27 | 49.2K | 67.8K | 7.90× | — |
Per task of the Artificial Analysis Intelligence Index: the average over its nine evaluations, counting input, cache, and output tokens at list prices. These are long, agent-style benchmark tasks, much longer than a chat message.
Where GPT-6.1 Sol sits on the frontier
Intelligence Index against one 100K-input, 10K-output request.