Users call GPT 5.6 Sol's stalling without repeated prompts insane or stupid because the system fails to continue tasks or run tests despite explicit instructions.
Based on 2 visible X reactions from 1 accounts; directional sample.
Ask a question below.
Published answers will appear here.
This is kind of insane - it told me it would continue, but it doesn't. It also wasn't checking in the work or running the tests - despite the agents.md explicitly telling it to check in work routinely, without user confirmation. https://x.com/davidmanheim/status/2076298306934710599/photo/1
...this is just stupid. https://x.com/davidmanheim/status/2076298960625438861/photo/1
When people said GPT 5.6 Sol was as good as Fable, but didn't run ahead as much (@petergostev/@mitchellh/@jayair/@deredleritt3r/@TheZvi), I didn't realize they meant it would just stop working and need handholding and repeated prompting to do anything further. Quite silly. https://x.com/davidmanheim/status/2076295803652456899/photo/1
This is kind of insane - it told me it would continue, but it doesn't. It also wasn't checking in the work or running the tests - despite the agents.md explicitly telling it to check in work routinely, without user confirmation. https://x.com/davidmanheim/status/2076298306934710599/photo/1
When people said GPT 5.6 Sol was as good as Fable, but didn't run ahead as much (@petergostev/@mitchellh/@jayair/@deredleritt3r/@TheZvi), I didn't realize they meant it would just stop working and need handholding and repeated prompting to do anything further. Quite silly. https://x.com/davidmanheim/status/2076295803652456899/photo/1
Users call GPT 5.6 Sol's stalling without repeated prompts insane or stupid because the system fails to continue tasks or run tests despite explicit instructions.
Based on 2 visible X reactions from 1 accounts; directional sample.
Ask a question below.
Published answers will appear here.