From cf667b5f3e05cbe84fe2d1a5ce159553f37f7a62 Mon Sep 17 00:00:00 2001
From: Asher Feldman <59994+asher@users.noreply.github.com>
Date: Sun, 9 Aug 2026 07:25:08 -0700
Subject: [PATCH] release 0.3.11

---
 CHANGELOG.md           | 12 ++++++++----
 mlx_kquant/_version.py |  2 +-
 2 files changed, 9 insertions(+), 5 deletions(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 81b326a..b22906f 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -6,6 +6,14 @@ adhere to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
 ## [Unreleased]
 
+## [0.3.11]
+
+### Changed
+- MoE prefill gather (sorted-rhs NAX path) is 12-28% faster per call
+  below ~64 rows per expert, biggest at 128-529-token chunks on top-8
+  256-expert shapes (bit-identical; KQ_GATHER_RHS_NAX_BM forces the
+  tile height).
+
 ## [0.3.10]
 
 ### Added
@@ -20,10 +28,6 @@ adhere to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
   fine buffers during deep prefill.
 
 ### Changed
-- MoE prefill gather (sorted-rhs NAX path) is 12-28% faster per call
-  below ~64 rows per expert, biggest at 128-529-token chunks on top-8
-  256-expert shapes (bit-identical; KQ_GATHER_RHS_NAX_BM forces the
-  tile height).
 - iq2_xxs / iq2_xs / iq2_s / iq3_s MoE gather decode is 9-12% faster per
   call (hoisted block scale, byte-indexed grids); the ext mat-vec at
   verify widths 2..8 gains 7-10% on the same codecs.
diff --git a/mlx_kquant/_version.py b/mlx_kquant/_version.py
index aa4cd15..91c67cc 100644
--- a/mlx_kquant/_version.py
+++ b/mlx_kquant/_version.py
@@ -1 +1 @@
-__version__ = "0.3.10"
+__version__ = "0.3.11"