From 43b5a57dbbf6b7ea859f9bdf492e57d3b86207bc Mon Sep 17 00:00:00 2001
From: zijiexia <37504505+zijiexia@users.noreply.github.com>
Date: Thu, 27 Aug 2026 21:56:10 -0700
Subject: [PATCH] [Docs] Feature GLM-5.3-Flash in the popular-models banner
 (#36784)

Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
---
 docs/cookbook/autoregressive/intro.mdx       |  2 +-
 docs/src/snippets/configs/popular-models.jsx | 34 ++++++++++----------
 2 files changed, 18 insertions(+), 18 deletions(-)

diff --git a/docs/cookbook/autoregressive/intro.mdx b/docs/cookbook/autoregressive/intro.mdx
index 86fc4abce..176293793 100644
--- a/docs/cookbook/autoregressive/intro.mdx
+++ b/docs/cookbook/autoregressive/intro.mdx
@@ -22,7 +22,7 @@ metatags:
   <Card
     title="GLM"
     mode="card"
-    href="/cookbook/autoregressive/GLM/GLM-5.2"
+    href="/cookbook/autoregressive/GLM/GLM-5.3-Flash"
     img="/cards/logos/glm.png"
   />
   <Card
diff --git a/docs/src/snippets/configs/popular-models.jsx b/docs/src/snippets/configs/popular-models.jsx
index 400d6574d..d2df38a71 100644
--- a/docs/src/snippets/configs/popular-models.jsx
+++ b/docs/src/snippets/configs/popular-models.jsx
@@ -29,6 +29,23 @@ export const popularModels = [
       caption: "Qwen3.8-Flash-Next deployment guide",
     },
   },
+  {
+    name: "GLM-5.3-Flash",
+    vendor: "Z.ai",
+    href: "/cookbook/autoregressive/GLM/GLM-5.3-Flash",
+    logo: "/cards/logos/glm.png",
+    badge: "New",
+    tags: ["9 platforms", "MLA + DSA + KDA hybrid", "Multimodal"],
+    hero: {
+      eyebrow: "Featured model · New",
+      headline: "Meet GLM-5.3-Flash on SGLang",
+      blurb:
+        "Z.ai's natively multimodal Mixture-of-Experts model — 320B total parameters with 18B active, 45 text layers combining MLA, DSA sparse, and KDA linear attention, a 24-layer vision encoder for image and video input, and a native MTP draft layer for speculative decoding. Recipes cover H100 / H200 / B200 / B300 / GB200 / GB300 and AMD MI300X / MI325X / MI355X.",
+      tags: ["320B / 18B active", "1M context", "Text + image + video"],
+      cta: "Open the GLM-5.3-Flash cookbook",
+      caption: "GLM-5.3-Flash deployment guide",
+    },
+  },
   {
     name: "MiniMax-H3",
     vendor: "MiniMax",
@@ -63,21 +80,4 @@ export const popularModels = [
       caption: "Kimi-K3 deployment guide",
     },
   },
-  {
-    name: "Inkling",
-    vendor: "Thinking Machines",
-    href: "/cookbook/autoregressive/ThinkingMachines/Inkling",
-    logo: "/cards/logos/thinkingmachines.png",
-    badge: "New",
-    tags: ["7 platforms", "NVFP4 / BF16", "MTP + DSpark"],
-    hero: {
-      eyebrow: "Featured model · New",
-      headline: "Meet Inkling on SGLang",
-      blurb:
-        "Thinking Machines' open-weights Mixture-of-Experts model — 975B parameters, 41B active per token, a 1M-token context window, and native text, image, and audio input. The cookbook covers its MTP speculative-decoding path and long-context prefix caching on NVIDIA and AMD.",
-      tags: ["975B · 41B active", "1M context", "Text + image + audio"],
-      cta: "Open the Inkling cookbook",
-      caption: "Inkling deployment guide",
-    },
-  },
 ];