gpt-5-6 (GPT-5.6 Sol) model/gpt-5-6

What it is

The model serving the measurement account. routing.model_slug is gpt-5-6 in standard mode and gpt-5-6-thinking in thinking mode; the UI calls it „GPT-5.6 Sol". Measured since 20 Aug 2026. On 21 Aug isolated gpt-5-5 answers still occurred, from 22 Aug only 5.6.

For visibility

The model is the single biggest context factor of every measurement. Search decision, fan-out phrasing and the format of the web.run calls hang on the generation — the pipe syntax, for instance, came back with a particular model phase. That is why the model_slug is recorded in every run's context snapshot and a change is tracked as a baseline event.

Change history

Open

• On 21 Aug two prompts still answered with gpt-5-5 (K4-02, K8-01), from 22 Aug consistently gpt-5-6. Rollout in waves or chance?
• How do gpt-5-6 and gpt-5-6-thinking differ in fan-out? Observation: thinking bundles more queries on product questions (up to 7) but barely thinks longer (2–6 s).

Related

Search decision web.run syntax (pipe format) Query fan-out