Multimodal reasoning model for visual analysis, planning, and tool use
Details
| Field | Value |
|---|---|
| id | baidu/ernie-4.5-vl-28b-a3b |
| modality | text+image->text |
| context_length | 30000 |
| max_output_tokens | 8000 |
| quantization | fp16 |
| release_date | 2025-06-30 |
| is_open_weights | yes |
| supports_tool_call | yes |
| supports_reasoning | yes |
| supports_structured_output | no |
| supports_attachment | yes |
| input_modalities | text, image |
| output_modalities | text |