GPT-6 Sol leaks, screenshots and release rumors

A developer says GPT-6 Sol generated ObsidianUI, a React and Tailwind CSS component library with MCP agents, in under 20 minutes, and shared a demo video.

A moonlit Three.js skateboard experience whose author stresses it was made with GPT-6 Sol, not Opus 5.5.

After several hours of testing, the user says Sol falls well short of Opus 5.5 on frontend and backend work, with generous usage as its only bright spot, and plans to return to Opus.

Given the same prompt, Opus 5.5, GPT-6 Sol, GPT-6 Luna and Grok 4.7 each built a launch site for a fictional handheld console, with time and cost compared.

A Chinese user says GPT-6 Sol can't even fix basic website UI code without repeated back-and-forth and feels dumber than 5.6 Sol, recommending Astra Extra High for complex tasks.

Given the same prompt, Opus 5.5 produced the best frontend and 3D detail at $8.95 and 10 minutes, while GPT-6 Sol was cheapest and fastest at $3.90 and 5 minutes but gave the most disappointing result.

Arena reports GPT-6 Sol (Max) placed #4 in Code Arena WebDev with 1689 points, matching Opus 5 (Max) at less than half the price, with large gains across all subcategories.

Testing four models on the same kanban app prompt, the author ranks Opus 5.5 first, Astra and Fable 5.1 tied, and GPT-6 Sol last for frontend design.

The same prompt asks both models to draw an animated pixel rabbit archer using only HTML and Canvas code, and a video compares the results.

Vercel's Rauch shared fresh Next.js evals where Opus 5.5, GPT-6 Sol and Fable 5.1 all score 97%, while Grok 4.7 scores 94% at 2x to 7x lower cost.

The same prompt had GPT-6 Sol and Opus 5 each build a robotic hummingbird in three.js; the author notes Sol is cheap with 50% off API pricing, but the comparison is against the older Opus 5.

GPT-6 Sol scores 97% on the Next.js evals leaderboard, tying Claude Opus 5.5 and Fable 5.1 for the highest success rate, with Opus having the best average cost of the three.