The report, posted on 15 October about an incident on 17 June, describes spoofed headers, path traversal, leaked server credentials and forged unsigned tokens — after the model had already found the number it was asked for through a public search.
Model
GPT-6 Sol
GPT-6 Sol is an artificial intelligence model developed by OpenAI. This page lists what is publicly known about the model and every benchmark result Aivio News has recorded for it, each with its own source and date.
- Developer
- OpenAI
- Model
- GPT
- Released
- September 22, 2026
- Modalities
- Model, Image, Code
- Status
- Active
- Source
- openai.com
Benchmarks
All stories| Benchmark | Score | Source | Evaluated |
|---|---|---|---|
| Vibe Code Bench single agent, medium effort | 77.6% | vals.ai Independent Verified | — Source date: October 9, 2026 |
| Vibe Code Bench single agent, maximum effort | 89% | vals.ai Independent Verified | — Source date: October 9, 2026 |
More from GPT-6 Sol
Agent teams cost up to 5.1 times as much as one agent, and beat it in only one of four comparisons
On Vals AI's Vibe Code Bench, a lead agent delegating to subagents beat a single agent once out of four — and simply raising the single agent's reasoning effort did more for less.
A month after GPT-6 launched for paying customers, OpenAI has put it in front of everyone — and changed what an answer looks like while it was at it.
OpenAI's new GPT-6.1 Sol comes within 2.1 points of Astra on computer use, at a seventh of the cost
The model went live hours after OpenAI confirmed it would not ship GPT-6.1 Astra at all, and it undercuts Anthropic's Opus 5.5 on two of the benchmarks OpenAI published.
DrivingBench puts frontier models in command of a real car's steering, throttle and brakes at walking pace. The gap between first and second place is the story.
Sol is aimed at coding and complex work, Luna at high-volume clerical tasks. OpenAI says the pair cost about half as much as the 5.6 generation they replace.