肖泽涛 0695f6f4cd fix(embedding): use model-level batch_size for provider-specific limits ... Root cause: DashScope embeddings accepts at most 10 inputs per request, but code used fixed batch_size=40, causing 400 Bad Request during indexing. Changes: - add batch_size to EmbedModelInfo with default 40 - set dashscope/text-embedding-v4 batch_size to 10 - persist batch_size in BaseEmbeddingModel and use it as default in batch/abatch encode - update Milvus, evaluation, and upload-graph embedding paths to read model-configured batch_size		2026-02-28 13:09:53 +08:00
..
static	fix(embedding): use model-level batch_size for provider-specific limits	2026-02-28 13:09:53 +08:00
__init__.py	lint: 代码格式化	2025-09-18 20:32:33 +08:00
app.py	feat(account): 添加部门基础管理功能 #480	2026-01-18 22:54:43 +08:00