core
API
Generated with
core/org.llm4s/org.llm4s.context/DeterministicCompressor

DeterministicCompressor

org.llm4s.context.DeterministicCompressor
object DeterministicCompressor

Implements rule-based deterministic compression for conversation context.

This compressor applies predictable, reproducible transformations to reduce token usage while preserving semantic meaning. Unlike LLM-based compression, the output is deterministic and doesn't require API calls.

==Compression Pipeline==

Compression occurs in two phases:

  1. '''Tool Compaction''' (always applied):
  • Compresses large JSON/YAML tool outputs
  • Externalizes binary content
  • Truncates verbose logs and error traces
  1. '''Subjective Edits''' (optional, requires enableSubjectiveEdits):
  • Removes filler words from transcript-like content
  • Deduplicates repetitive sentences
  • Truncates overly verbose assistant responses

==Safety Guarantees==

The compressor is designed to be '''safe''' and '''conservative''':

  • User messages are '''never''' modified
  • Code blocks and JSON are preserved verbatim
  • Filler word removal only applies to "transcript-like" content
  • Truncation preserves first and last sentences

Attributes

See also

ToolOutputCompressor for the tool compaction implementation

CompressionRule for individual compression rules

Example 
val compressed = DeterministicCompressor.compressToCap(
 messages = conversation.messages,
 tokenCounter = counter,
 capTokens = 4000,
 enableSubjectiveEdits = true
)
Graph
Supertypes
class Object
trait Matchable
class Any
Self type

Members list

Value members

Concrete methods

def compressToCap(messages: Seq[Message], tokenCounter: ConversationTokenCounter, capTokens: Int, artifactStore: Option[ArtifactStore], enableSubjectiveEdits: Boolean): Result[Seq[Message]]

Compress messages to fit within a token budget.

Compress messages to fit within a token budget.

Applies tool compaction first, then optionally subjective edits until the total token count is at or below capTokens.

Value parameters

artifactStore

Optional store for externalized content (defaults to in-memory)

capTokens

Target maximum token count

enableSubjectiveEdits

Whether to apply filler word removal and truncation

messages

The messages to compress

tokenCounter

Token counter for measuring compression progress

Attributes

Returns

Compressed messages, or an error if compression fails

Generated with