Session memory¶

ConversationMemory accumulates a turn-by-turn message history and applies a compaction strategy after each append, so a long session stays within the model's context window. Memory is caller-driven: append what you send and receive, then read messages() to build the next prompt.

from techrevati.runtime import (
    InMemoryConversationMemory, MemoryMessage, TokenBudgetCompaction,
)

memory = InMemoryConversationMemory(
    compaction=TokenBudgetCompaction(max_tokens=8000),
)
memory.add(MemoryMessage("system", "You are a loan assistant."))
memory.add(MemoryMessage("user", question))

reply, _ = session.run_turn(lambda: call_model(memory.messages()))
memory.add(MemoryMessage("assistant", reply))

Compaction strategies:

• NoCompaction — keep everything (default).
• WindowCompaction(max_messages=..., keep_system=True) — keep the last N messages; system messages are retained first.
• TokenBudgetCompaction(max_tokens=..., estimator=...) — drop oldest messages until an estimated token budget is met. System messages are retained first; if they alone exceed the budget they are all kept — correctness over budget, never silently dropping instructions.

Write a custom strategy by implementing CompactionStrategy.compact.