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

docs: remove DSV4 low-latency chunked prefill size (#34333)

Browse Source
This commit is contained in:
Baizhou Zhang
2026-08-10 17:23:43 -07:00
committed by GitHub
parent d59c1ddf70
commit a92bbf2f24
1 changed files with 0 additions and 11 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/src/snippets/configs/deepseek-ai/deepseek-v4.jsx
-11
View File
@@ -450,7 +450,6 @@ sgl-eval run aime25 \\
"--tp 8",
"--moe-runner-backend flashinfer_mxfp4",
"--speculative-algorithm DSPARK",
"--chunked-prefill-size 4096",
"--disable-flashinfer-autotune",
"--swa-full-tokens-ratio 0.1",
"--mem-fraction-static 0.90",
@@ -471,7 +470,6 @@ sgl-eval run aime25 \\
"--speculative-num-steps 3",
"--speculative-eagle-topk 1",
"--speculative-num-draft-tokens 4",
"--chunked-prefill-size 4096",
"--disable-flashinfer-autotune",
"--swa-full-tokens-ratio 0.1",
"--host {{HOST_IP}}",
@@ -646,7 +644,6 @@ sgl-eval run aime25 \\
"--speculative-num-steps 3",
"--speculative-eagle-topk 1",
"--speculative-num-draft-tokens 4",
"--chunked-prefill-size 4096",
"--disable-flashinfer-autotune",
"--swa-full-tokens-ratio 0.1",
"--host {{HOST_IP}}",
@@ -830,7 +827,6 @@ sgl-eval run aime25 \\
"--speculative-num-steps 3",
"--speculative-eagle-topk 1",
"--speculative-num-draft-tokens 4",
"--chunked-prefill-size 4096",
"--disable-flashinfer-autotune",
"--swa-full-tokens-ratio 0.1",
"--host {{HOST_IP}}",
@@ -853,7 +849,6 @@ sgl-eval run aime25 \\
"--speculative-num-steps 3",
"--speculative-eagle-topk 1",
"--speculative-num-draft-tokens 4",
"--chunked-prefill-size 4096",
"--disable-flashinfer-autotune",
"--swa-full-tokens-ratio 0.1",
"--host {{HOST_IP}}",
@@ -895,7 +890,6 @@ sgl-eval run aime25 \\
"--tp 4",
"--moe-runner-backend flashinfer_mxfp4",
"--speculative-algorithm DSPARK",
"--chunked-prefill-size 4096",
"--disable-flashinfer-autotune",
"--swa-full-tokens-ratio 0.1",
"--host {{HOST_IP}}",
@@ -915,7 +909,6 @@ sgl-eval run aime25 \\
"--speculative-num-steps 3",
"--speculative-eagle-topk 1",
"--speculative-num-draft-tokens 4",
"--chunked-prefill-size 4096",
"--disable-flashinfer-autotune",
"--swa-full-tokens-ratio 0.1",
"--host {{HOST_IP}}",
@@ -1082,7 +1075,6 @@ sgl-eval run aime25 \\
"--speculative-num-steps 3",
"--speculative-eagle-topk 1",
"--speculative-num-draft-tokens 4",
"--chunked-prefill-size 4096",
"--disable-flashinfer-autotune",
"--swa-full-tokens-ratio 0.1",
"--host {{HOST_IP}}",
@@ -1125,7 +1117,6 @@ sgl-eval run aime25 \\
"--moe-runner-backend flashinfer_mxfp4",
"--speculative-algorithm DSPARK",
"--mem-fraction-static 0.90",
"--chunked-prefill-size 4096",
"--swa-full-tokens-ratio 0.1",
"--host {{HOST_IP}}",
"--port {{PORT}}",
@@ -1144,7 +1135,6 @@ sgl-eval run aime25 \\
"--speculative-num-steps 3",
"--speculative-eagle-topk 1",
"--speculative-num-draft-tokens 4",
"--chunked-prefill-size 4096",
"--swa-full-tokens-ratio 0.1",
"--host {{HOST_IP}}",
"--port {{PORT}}",
@@ -1300,7 +1290,6 @@ sgl-eval run aime25 \\
"--speculative-num-steps 3",
"--speculative-eagle-topk 1",
"--speculative-num-draft-tokens 4",
"--chunked-prefill-size 4096",
"--disable-flashinfer-autotune",
"--swa-full-tokens-ratio 0.1",
"--host {{HOST_IP}}",