The text provider now includes reference_images alongside inputs when building prompts. Image files are sent as base64 data URLs via ImageURLChunk for actual multimodal vision support, replacing the previous [Attached image: ...] placeholder text. |
||
|---|---|---|
| .. | ||
| providers | ||
| __init__.py | ||
| builder.py | ||
| cli.py | ||
| config.py | ||
| graph.py | ||
| state.py | ||