基于 Qwen3.7 训练的轻量级多语言文本向量模型,面向成本和吞吐敏感场景。相较 text-embedding-v4,在多语言覆盖(100→201 种)、长文本处理(32K→128K)和 Sparse Embedding 等能力上大幅升级;在语义、跨语言及代码检索评测上整体效果基本持平,其中 AIRBench-zh 提升约 3%
curl https://api.yourplatform.com/v1/chat/completions \
-H "Content-Type: application/json" \
-H "Authorization: Bearer YOUR_API_KEY" \
-d '{
"model": "undefined",
"messages": [
{"role": "user", "content": "Hello!"}
],
"stream": true
}'