minke.yu/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 12 Packages Projects Releases Wiki Activity

docs: split MI300X and MI325X options in GLM-5.1 generator (#23540)

Browse Source 
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
This commit is contained in:
zijiexia
2026-04-23 12:01:58 -07:00
committed by GitHub
co-authored by gemini-code-assist[bot]
parent 80125febb1
commit 9b2f7f8a91
3 changed files with 15 additions and 13 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs_new/cookbook/autoregressive/intro.mdx
+1 -1
View File
@@ -28,7 +28,7 @@ metatags:
<Card <Card
title="GLM" title="GLM"
mode="card" mode="card"
href="/cookbook/autoregressive/GLM/GLM-4.5" href="/cookbook/autoregressive/GLM/GLM-5.1"
img="/cards/logos/glm.png" img="/cards/logos/glm.png"
/> />
<Card <Card
docs_new/docs.json
+8 -8
View File
@@ -960,16 +960,16 @@
{ {
"group": "GLM", "group": "GLM",
"pages": [ "pages": [
"cookbook/autoregressive/GLM/GLM-4.5", "cookbook/autoregressive/GLM/GLM-5.1",
"cookbook/autoregressive/GLM/GLM-4.6", "cookbook/autoregressive/GLM/GLM-5",
"cookbook/autoregressive/GLM/GLM-OCR",
"cookbook/autoregressive/GLM/GLM-Glyph",
"cookbook/autoregressive/GLM/GLM-4.7", "cookbook/autoregressive/GLM/GLM-4.7",
"cookbook/autoregressive/GLM/GLM-4.7-Flash", "cookbook/autoregressive/GLM/GLM-4.7-Flash",
"cookbook/autoregressive/GLM/GLM-5", "cookbook/autoregressive/GLM/GLM-4.6",
"cookbook/autoregressive/GLM/GLM-5.1", "cookbook/autoregressive/GLM/GLM-4.6V",
"cookbook/autoregressive/GLM/GLM-Glyph", "cookbook/autoregressive/GLM/GLM-4.5",
"cookbook/autoregressive/GLM/GLM-OCR", "cookbook/autoregressive/GLM/GLM-4.5V"
"cookbook/autoregressive/GLM/GLM-4.5V",
"cookbook/autoregressive/GLM/GLM-4.6V"
] ]
}, },
{ {
docs_new/src/snippets/autoregressive/glm-51-deployment.jsx
+6 -4
View File
@@ -14,7 +14,8 @@ export const GLM51Deployment = () => {
{ id: 'b200', label: 'B200', default: false }, { id: 'b200', label: 'B200', default: false },
{ id: 'gb300', label: 'GB300', default: false }, { id: 'gb300', label: 'GB300', default: false },
{ id: 'h100', label: 'H100', default: false }, { id: 'h100', label: 'H100', default: false },
{ id: 'mi300x', label: 'MI300X/MI325X', default: false }, { id: 'mi300x', label: 'MI300X', default: false },
{ id: 'mi325x', label: 'MI325X', default: false },
{ id: 'mi355x', label: 'MI355X', default: false } { id: 'mi355x', label: 'MI355X', default: false }
] ]
}, },
@@ -23,7 +24,7 @@ export const GLM51Deployment = () => {
title: 'Quantization', title: 'Quantization',
getDynamicItems: (values) => { getDynamicItems: (values) => {
const hw = values.hardware; const hw = values.hardware;
const isAMD = hw === 'mi300x' || hw === 'mi355x'; const isAMD = ['mi300x', 'mi325x', 'mi355x'].includes(hw);
const isGB300 = hw === 'gb300'; const isGB300 = hw === 'gb300';
return [ return [
{ id: 'bf16', label: 'BF16', subtitle: 'Full Weights', default: isAMD, disabled: isGB300, disabledReason: isGB300 ? 'BF16 is not recommended on GB300 for GLM-5.1' : '' }, { id: 'bf16', label: 'BF16', subtitle: 'Full Weights', default: isAMD, disabled: isGB300, disabledReason: isGB300 ? 'BF16 is not recommended on GB300 for GLM-5.1' : '' },
@@ -58,7 +59,7 @@ export const GLM51Deployment = () => {
speculative: { speculative: {
name: 'speculative', name: 'speculative',
title: 'Speculative Decoding', title: 'Speculative Decoding',
condition: (values) => values.hardware !== 'mi300x' && values.hardware !== 'mi355x', condition: (values) => !['mi300x', 'mi325x', 'mi355x'].includes(values.hardware),
items: [ items: [
{ id: 'disabled', label: 'Disabled', default: false }, { id: 'disabled', label: 'Disabled', default: false },
{ id: 'enabled', label: 'Enabled', default: true } { id: 'enabled', label: 'Enabled', default: true }
@@ -72,6 +73,7 @@ export const GLM51Deployment = () => {
b200: { fp8: { tp: 8, mem: 0.9 }, bf16: { tp: 16, mem: 0.9 } }, b200: { fp8: { tp: 8, mem: 0.9 }, bf16: { tp: 16, mem: 0.9 } },
gb300: { fp8: { tp: 4, mem: 0.9 } }, gb300: { fp8: { tp: 4, mem: 0.9 } },
mi300x: { bf16: { tp: 8, mem: 0.80 } }, mi300x: { bf16: { tp: 8, mem: 0.80 } },
mi325x: { bf16: { tp: 8, mem: 0.80 } },
mi355x: { bf16: { tp: 8, mem: 0.80 } } mi355x: { bf16: { tp: 8, mem: 0.80 } }
}; };
@@ -129,7 +131,7 @@ export const GLM51Deployment = () => {
const generateCommand = () => { const generateCommand = () => {
const { hardware, quantization } = values; const { hardware, quantization } = values;
const isAMD = hardware === 'mi300x' || hardware === 'mi355x'; const isAMD = ['mi300x', 'mi325x', 'mi355x'].includes(hardware);
const isGB300 = hardware === 'gb300'; const isGB300 = hardware === 'gb300';
const effectiveQuant = isAMD ? 'bf16' : (isGB300 && quantization === 'bf16' ? 'fp8' : quantization); const effectiveQuant = isAMD ? 'bf16' : (isGB300 && quantization === 'bf16' ? 'fp8' : quantization);
const suffix = effectiveQuant === 'fp8' ? '-FP8' : ''; const suffix = effectiveQuant === 'fp8' ? '-FP8' : '';