Added GPT-6.1 Sol

OpenAI's GPT-6.1 Sol is now in the gallery. It produced HTML for all 35 prompts, and every session ended with a normal model stop.

The model used the pi harness to run its HTML, inspect the browser, and iterate before finishing. Sessions had a 15-minute deadline and a finalization reminder at 10 minutes, with no turn or browser-call cap. The batch used a rolling pool of up to 20 jobs. There is no judging or star rating.

Recorded model usage for the batch cost $5.24, excluding infrastructure. Explore the model page or open the featured builds below for their output, transcript, runtime, and cost.

Fireworks and Fluid Simulation generated successfully but initially timed out during final screenshot capture. Their previews were recovered separately from the saved HTML, without another model call.