Skip to content

Self-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills.

emb token (#1038)

yeshion23333ea240a22a164
diff --git a/openviking/session/session.py b/openviking/session/session.py
index f62998c99..aeb3f5bb6 100644
--- a/openviking/session/session.py
+++ b/openviking/session/session.py
@@ -81,6 +81,11 @@ class SessionMeta:
"total_tokens": 0,
}
)
+ embedding_token_usage: Dict[str, int] = field(
+ default_factory=lambda: {
+ "total_tokens": 0,
+ }
+ )
def to_dict(self) -> Dict[str, Any]:
return {
@@ -92,12 +97,15 @@ class SessionMeta:
"memories_extracted": dict(self.memories_extracted),
"last_commit_at": self.last_commit_at,
"llm_token_usage": dict(self.llm_token_usage),
+ "embedding_token_usage": dict(self.embedding_token_usage),
}
@classmethod
def from_dict(cls, data: Dict[str, Any]) -> "SessionMeta":
- token_usage = data.get("llm_token_usage", {})
+ llm_token_usage = data.get("llm_token_usage", {})
+ embedding_token_usage = data.get("embedding_token_usage", {})
memories = data.get("memories_extracted", {})
+
return cls(
session_id=data.get("session_id", ""),
created_at=data.get("created_at", ""),
@@ -117,9 +125,12 @@ class SessionMeta:
},
last_commit_at=data.get("last_commit_at", ""),
llm_token_usage={
- "prompt_tokens": token_usage.get("prompt_tokens", 0),
- "completion_tokens": token_usage.get("completion_tokens", 0),
- "total_tokens": token_usage.get("total_tokens", 0),
+ "prompt_tokens": llm_token_usage.get("prompt_tokens", 0),
+ "completion_tokens": llm_token_usage.get("completion_tokens", 0),
+ "total_tokens": llm_token_usage.get("total_tokens", 0),
+ },
+ embedding_token_usage={
+ "total_tokens": embedding_token_usage.get("total_tokens", 0),
},
)
@@ -551,6 +562,9 @@ class Session:
self._meta.llm_token_usage["prompt_tokens"] += llm.get("input", 0)
self._meta.llm_token_usage["completion_tokens"] += llm.get("output", 0)
self._meta.llm_token_usage["total_tokens"] += llm.get("total", 0)
+
+ embedding = snapshot.summary.get("tokens", {}).get("embedding", {})
+ self._meta.embedding_token_usage["total_tokens"] += embedding.get("total", 0)
self._meta.commit_count = self._compression.compression_index
for cat, count in memories_extracted.items():
self._meta.memories_extracted[cat] = (
@@ -573,6 +587,13 @@ class Session:
"archive_uri": archive_uri,
"memories_extracted": memories_extracted,
"active_count_updated": active_count_updated,
+ "token_usage": {
+ "llm": dict(self._meta.llm_token_usage),
+ "embedding": dict(self._meta.embedding_token_usage),
+ "total": {
+ "total_tokens": self._meta.llm_token_usage["total_tokens"] + self._meta.embedding_token_usage["total_tokens"]
+ }
+ }
},
)
logger.info(f"Session {self.session_id} memory extraction completed")