Cohere vision model for multilingual document analysis, OCR, and image understanding
Details
| Field | Value |
|---|---|
| id | command-a-vision-07-2025 |
| family | command-a |
| modality | text+image->text |
| context_length | 128000 |
| max_output_tokens | 8000 |
| knowledge_cutoff | 2024-06-01 |
| release_date | 2025-07-31 |
| is_open_weights | yes |
| supports_tool_call | no |
| supports_reasoning | no |
| supports_structured_output | no |
| supports_attachment | yes |
| input_modalities | text, image |
| output_modalities | text |