From 543eca3d9633ef78ff5ec8727d4bd1a47ab9f292 Mon Sep 17 00:00:00 2001 From: Simba Zhang Date: Sun, 16 Aug 2026 07:32:44 -0700 Subject: [PATCH] chore: bump mlx-swift-lm to pick up MTP head quantization fix MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fixes #153. mlx-swift-lm#56 fixes a per-layer quantization lookup bug where MTP prediction-head modules (indexed as mtp..layers... in the Swift module tree, but keyed as mtp.layers... in checkpoint quantization overrides) fell through to the top-level default quantization instead of their own override — crashing quantized_matmul on checkpoints with a separately-quantized MTP head, e.g. a uniform 8-bit MTP head over a mixed 4/5/6-bit qwen3_5_moe main model, under --stream-experts --mtp. --- mlx-swift-lm | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mlx-swift-lm b/mlx-swift-lm index 512dcde..a34a424 160000 --- a/mlx-swift-lm +++ b/mlx-swift-lm @@ -1 +1 @@ -Subproject commit 512dcded41f940b9111c7e69d1642ff7bd0916ae +Subproject commit a34a424bd17f412e3343f44f080bf57dc96a7d4f