youngkingdom/dify
1
0
Fork 0
mirror of https://github.com/langgenius/dify.git synced 2026-05-01 07:58:02 +08:00
Code Issues Packages Projects Releases Wiki Activity

feat(telemetry): add input/output token split to enterprise OTEL traces

Browse Source 
- Add PROMPT_TOKENS and COMPLETION_TOKENS to WorkflowNodeExecutionMetadataKey
- Store prompt/completion tokens in node execution metadata JSON (no schema change)
- Calculate workflow-level token split by summing node executions on-the-fly
- Export gen_ai.usage.input_tokens and output_tokens to enterprise telemetry
- Add semantic convention constants for token attributes
- Maintain backward compatibility (historical data shows null)

BREAKING: None
MIGRATION: None (uses JSON metadata, no schema changes)
This commit is contained in:
GareArc
2026-02-03 19:27:11 -08:00
parent 701f02f853
commit 8ceb1ed96f
8 changed files with 56 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
api/core/app/workflow/layers/persistence.py
View File

@ -439,6 +439,8 @@ class WorkflowPersistenceLayer(GraphEngineLayer):
"created_at": domain_execution.created_at,
"finished_at": domain_execution.finished_at,
"total_tokens": meta.get(WorkflowNodeExecutionMetadataKey.TOTAL_TOKENS, 0),
"prompt_tokens": meta.get(WorkflowNodeExecutionMetadataKey.PROMPT_TOKENS),
"completion_tokens": meta.get(WorkflowNodeExecutionMetadataKey.COMPLETION_TOKENS),
"total_price": meta.get(WorkflowNodeExecutionMetadataKey.TOTAL_PRICE, 0.0),
"currency": meta.get(WorkflowNodeExecutionMetadataKey.CURRENCY),
"tool_name": (meta.get(WorkflowNodeExecutionMetadataKey.TOOL_INFO) or {}).get("tool_name")