| fix(model): default custom providers to image input
Allow newly configured OpenAI/Anthropic-compatible models to read images without requiring per-model multimodal YAML.
Co-authored-by: Cursor <cursoragent@cursor.com>
| 15 天前 |
| Drop orphaned OpenAI tool results
Co-authored-by: Cursor <cursoragent@cursor.com>
| 5 天前 |
| Fix OpenAI tool call normalization
Co-authored-by: Cursor <cursoragent@cursor.com>
| 5 天前 |
| Fix OpenAI tool call normalization
Co-authored-by: Cursor <cursoragent@cursor.com>
| 5 天前 |