Documentation Index
Fetch the complete documentation index at: https://mintlify.com/jxnl/kura/llms.txt
Use this file to discover all available pages before exploring further.
SummaryModel
Instructor-based summary model for conversation analysis using the CLIO framework.
Constructor
SummaryModel(
model: Union[str, "KnownModelName"] = "openai/gpt-4o-mini",
max_concurrent_requests: int = 50,
checkpoint_filename: str = "summaries",
console: Optional[Console] = None,
cache: Optional[CacheStrategy] = None,
)
model
Union[str, KnownModelName]
default:"openai/gpt-4o-mini"
Model identifier (e.g., “openai/gpt-4o-mini”, “anthropic/claude-3-5-sonnet”)
Maximum concurrent API requests
Filename for checkpointing this model’s output
console
Optional[Console]
default:"None"
Rich console for progress tracking
cache
Optional[CacheStrategy]
default:"None"
Caching strategy to use (optional)
Methods
summarise()
Summarise conversations with configurable parameters using the CLIO framework.
async def summarise(
conversations: list[Conversation],
prompt: str = DEFAULT_SUMMARY_PROMPT,
*,
response_schema: Type[T] = GeneratedSummary,
temperature: float = 0.2,
**kwargs,
) -> list[ConversationSummary]
conversations
list[Conversation]
required
List of conversations to summarize
prompt
str
default:"DEFAULT_SUMMARY_PROMPT"
Custom prompt for CLIO analysis
response_schema
Type[T]
default:"GeneratedSummary"
Pydantic model class for structured LLM output. Extend GeneratedSummary to add custom fields that will automatically be included in ConversationSummary.metadata
LLM temperature for generation
return
list[ConversationSummary]
List of ConversationSummary objects with core fields populated and any additional fields from extended schemas in metadata
from kura.summarisation import SummaryModel
from kura.types import Conversation
from kura.types.summarisation import GeneratedSummary
class CustomSummary(GeneratedSummary):
sentiment: str
technical_complexity: int
model = SummaryModel()
summaries = await model.summarise(
conversations,
response_schema=CustomSummary,
prompt="Rate sentiment and technical complexity 1-10"
)
# Access core fields
print(summaries[0].summary)
# Access custom fields in metadata
print(summaries[0].metadata["sentiment"])
summarise_conversations()
Generate summaries for a list of conversations using the CLIO framework. This is a pure function that takes conversations and a summary model, and returns conversation summaries with automatic extensibility.
async def summarise_conversations(
conversations: list[Conversation],
*,
model: BaseSummaryModel,
response_schema: Type[T] = GeneratedSummary,
prompt: str = DEFAULT_SUMMARY_PROMPT,
temperature: float = 0.2,
checkpoint_manager: Optional[BaseCheckpointManager] = None,
**kwargs,
) -> list[ConversationSummary]
conversations
list[Conversation]
required
List of conversations to summarize
Model to use for summarization (OpenAI, vLLM, local, etc.)
response_schema
Type[T]
default:"GeneratedSummary"
Pydantic model class for LLM output. Extend GeneratedSummary to add custom fields that will appear in metadata
prompt
str
default:"DEFAULT_SUMMARY_PROMPT"
Custom prompt to modify the CLIO analysis
LLM temperature for generation
checkpoint_manager
Optional[BaseCheckpointManager]
default:"None"
Optional checkpoint manager for caching
return
list[ConversationSummary]
List of ConversationSummary objects with core CLIO fields and any additional fields from extended schemas in metadata
from kura.summarisation import SummaryModel, summarise_conversations
model = SummaryModel()
summaries = await summarise_conversations(
conversations=my_conversations,
model=model
)
from kura.summarisation import summarise_conversations
from kura.types.summarisation import GeneratedSummary
class DetailedSummary(GeneratedSummary):
sentiment: str
technical_depth: int
summaries = await summarise_conversations(
conversations=my_conversations,
model=model,
response_schema=DetailedSummary,
prompt="Analyze sentiment and rate technical depth 1-10"
)
# Custom fields available in metadata
print(summaries[0].metadata["sentiment"])