GPT-6 Sol leaks, screenshots and release rumors

A leak claims Sonnet 5.5 will launch tomorrow at 2 PM New York time, is in its final stages, slightly beats GPT-6 Sol in early testing, comes close to Opus 5.5 in capability but cheaper, and a claude-sonnet-5-5 model tag has already appeared.

OpenAI's changelog confirms an image encoding bug had lowered Sol and Luna image understanding across the API, Codex vision tasks and Computer Use; Luna's ReCoG visual grounding rose from 28.8% to 60.0% after the fix, and OpenAI advises rerunning evaluations for affected workflows.

A video post shows GPT-6 Sol and Claude Opus 5.5 being tested together on 3D tasks in Unreal Engine and Blender.

mark_k argues OpenAI is short on GPU capacity right now, so GPT-6 Sol is likely a smaller model replacing Terra to relieve inference load, which explains its lackluster performance.

A Japanese-language morning roundup covers several AI developments, including a fix for an image understanding bug in GPT-6 Sol and Luna and a reset of usage limits.

A hands-on comparison video has Opus 5.5 and GPT-6 Sol each complete a Premiere Pro editing task to see which model handles real software better.

A poster says testing on the day showed GPT-6 Sol starting to refuse offensive security research requests, including authorized security research.

The author, who has built FPS demos and generated nearly 100 FPS videos, says the demo is AI-generated video passed off as gameplay and asks for the real Unreal project and a live control demo, noting credits are sold in the replies.

andonlabs reran Blueprint-Bench for GPT-6 Sol and Luna because of an OpenAI bug: Sol improved slightly while Luna made a large jump, with score charts included.

A long post argues that GPT-6 Sol's half-price launch still didn't bring developers back, since what matters is subscription allowance rather than unit price, and DevDay is unlikely to turn it around.

banteg says Opus 5.5 solved an 1,800-line player_update function, after over 100 agent-hours since July, including 50 hours from gpt-5.6-sol and 20 from gpt-6-astra.

On TapNow, the author gave GPT-6 Sol a story log and style reference images, and it handled the script, character and location design, and video prompts, with images generated via GPT Image 2.5 and MiniMax H3.