fix: stop passing est_tokens off as measured usage

usage 帧缺失/非法时不再拿 est_tokens(最坏情形上界)当实测值,chat 与
embedding 两处兜底改记 0 并标 unavailable;打捞覆盖加 measured 前置条件,
避免 0/0 被洗成 estimated 而算出假的 0.0。embedding 全批合并扩三态(任一批
不可得 → 整体不可得),_total_cost 遇不可得批整体记 NULL。
This commit is contained in:
2026-07-30 10:39:32 -04:00
parent 42e429eb58
commit 195454d2e3
6 changed files with 252 additions and 25 deletions
+1 -1
View File
@@ -284,7 +284,7 @@ class EmbeddingTransportResult:
vectors: list[list[float]]
dim: int
prompt_tokens: int
usage_source: str # measured | estimated
usage_source: str # measured | estimated | unavailable
raw: dict[str, Any]