From a92bbf2f247f4ec1f930193e266ec8392482da13 Mon Sep 17 00:00:00 2001 From: Baizhou Zhang Date: Mon, 10 Aug 2026 17:23:43 -0700 Subject: [PATCH] docs: remove DSV4 low-latency chunked prefill size (#34333) --- docs/src/snippets/configs/deepseek-ai/deepseek-v4.jsx | 11 ----------- 1 file changed, 11 deletions(-) diff --git a/docs/src/snippets/configs/deepseek-ai/deepseek-v4.jsx b/docs/src/snippets/configs/deepseek-ai/deepseek-v4.jsx index 0ef90141d..da33283ec 100644 --- a/docs/src/snippets/configs/deepseek-ai/deepseek-v4.jsx +++ b/docs/src/snippets/configs/deepseek-ai/deepseek-v4.jsx @@ -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}}",