doubao-embedding-vision · Doubao · embedding
豆包多模态向量化模型，支持图像和文本双模态输入。
Multimodal embedding (image + text); semantic search; RAG retrieval; clustering; classification.
Multimodal RAG; visual search; document clustering; image-text similarity matching.
No text generation; no Function Calling; output dimension fixed; image input costs more.
Anthropic 官方 CLI 编码智能体，支持多文件编辑、工具调用、长上下文推理。
export ANTHROPIC_AUTH_TOKEN="YOUR_API_KEY"
export ANTHROPIC_BASE_URL="/api/plan"
export ANTHROPIC_MODEL="doubao-embedding-vision"
claude --dangerously-skip-permissions