diff --git a/navi/config.py b/navi/config.py index 095b08d..b486674 100644 --- a/navi/config.py +++ b/navi/config.py @@ -119,7 +119,7 @@ # Context compression context_compression_enabled: bool = True - context_compression_threshold: float = 0.70 # trigger at 70% of ollama_num_ctx + context_compression_threshold: float = 0.90 # trigger at 90% of ollama_num_ctx context_keep_recent: int = 8 # conversational turns to keep verbatim context_summary_temperature: float = 0.3 context_summary_max_tokens: int = 4000 # max output tokens for the summary LLM call diff --git a/tests/unit/workers/test_compression_worker.py b/tests/unit/workers/test_compression_worker.py index 11f69ae..a407b32 100644 --- a/tests/unit/workers/test_compression_worker.py +++ b/tests/unit/workers/test_compression_worker.py @@ -15,7 +15,7 @@ def _ctx(session_id: str, llm, store) -> WorkerContext: return WorkerContext( session_id=session_id, - context_tokens=50_000, # over the 0.70 threshold -> worker attempts compression + context_tokens=60_000, # over the 0.90 threshold -> worker attempts compression max_context_tokens=65536, llm=llm, model="test",