Skip to content

feat: support passing cached_token #2

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
MengsD merged 7 commits into from
Jan 15, 2026
+21 −0
Merged

feat: support passing cached_token #2

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
a21e5b6
feat: support passing `cached_token` from dash call
MengsD Jan 14, 2026
718f729
remove typos
MengsD Jan 14, 2026
6110fee
fix typos
MengsD Jan 14, 2026
e573c61
fix typos
MengsD Jan 14, 2026
4ce6aeb
edited to explicitly pass cached_token
MengsD Jan 15, 2026
0ab5261
edited to explicitly pass cached_token
MengsD Jan 15, 2026
220fa4c
edited comments
MengsD Jan 15, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 8 additions & 0 deletions src/agentscope/model/_dashscope_model.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -368,6 +368,10 @@ async def _parse_dashscope_stream_response(
input_tokens=chunk.usage.input_tokens,
output_tokens=chunk.usage.output_tokens,
time=(datetime.now() - start_datetime).total_seconds(),
prompt_tokens_details=chunk.usage.get(
"prompt_tokens_details",
None,
),
)

parsed_chunk = ChatResponse(
Expand Down Expand Up @@ -469,6 +473,10 @@ async def _parse_dashscope_generation_response(
input_tokens=response.usage.input_tokens,
output_tokens=response.usage.output_tokens,
time=(datetime.now() - start_datetime).total_seconds(),
prompt_tokens_details=response.usage.get(
"prompt_tokens_details",
None,
),
)

parsed_response = ChatResponse(
Expand Down
3 changes: 3 additions & 0 deletions src/agentscope/model/_model_usage.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,5 +19,8 @@ class ChatUsage(DictMixin):
time: float
"""The time used in seconds."""

prompt_tokens_details: dict = field(default_factory=dict)
"""The details of cached tokens. """

type: Literal["chat"] = field(default_factory=lambda: "chat")
"""The type of the usage, must be `chat`."""
3 changes: 3 additions & 0 deletions src/agentscope/tracing/_attributes.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -68,6 +68,9 @@ class SpanAttributes:
GEN_AI_USAGE_OUTPUT_TOKENS = GenAIAttributes.GEN_AI_USAGE_OUTPUT_TOKENS
"""The gen ai usage output tokens."""

GEN_AI_USAGE_CACHED_TOKENS = "gen_ai.dash.usage.cached_tokens"
"""The gen ai usage cached tokens."""

# GenAI Message Attributes
GEN_AI_INPUT_MESSAGES = GenAIAttributes.GEN_AI_INPUT_MESSAGES
"""The gen ai input messages."""
Expand Down
7 changes: 7 additions & 0 deletions src/agentscope/tracing/_extractor.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -378,6 +378,13 @@ def _get_llm_response_attributes(
attributes[
SpanAttributes.GEN_AI_USAGE_OUTPUT_TOKENS
] = chat_response.usage.output_tokens
if (
hasattr(chat_response.usage, "prompt_tokens_details")
and chat_response.usage.prompt_tokens_details
):
attributes[
SpanAttributes.GEN_AI_USAGE_CACHED_TOKENS
] = _serialize_to_str(chat_response.usage.prompt_tokens_details)

output_messages = _get_llm_output_messages(chat_response)
if output_messages:
Expand Down
Loading