From d484de839073338b1ae0e1f1604a967c2fb22ba2 Mon Sep 17 00:00:00 2001 From: emil Date: Sat, 16 May 2026 14:10:01 +0300 Subject: [PATCH] =?UTF-8?q?Bump=20default=20n=5Fctx=208192=20=E2=86=92=203?= =?UTF-8?q?2768?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 8K filled up after one month of activity (~200 MB state) and the next `memwalk update` crashed with llama_decode returned 1 because new tokens didn't fit. Nemotron 3 Nano 4B is trained for 1M context so 32K is well within model capacity and leaves headroom for several more months of ingest before rotation matters. Note: existing users need to delete current.memb after upgrading, since state files are tied to the n_ctx they were created with. Co-Authored-By: Claude Opus 4.7 --- memwalk/config.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/memwalk/config.py b/memwalk/config.py index 4ecad40..ddd800c 100644 --- a/memwalk/config.py +++ b/memwalk/config.py @@ -49,7 +49,7 @@ class BashConfig: class Config: model_path: Path n_gpu_layers: int = -1 - n_ctx: int = 8192 + n_ctx: int = 32768 state_dir: Path = HOME_DIR git: GitConfig = field(default_factory=GitConfig) bash: BashConfig = field(default_factory=BashConfig)