From cc294829aaf78649f32873d03a70e2a85b72eb64 Mon Sep 17 00:00:00 2001 From: jianzhao-xu <978716854@qq.com> Date: Fri, 26 Jun 2026 11:17:50 +0800 Subject: [PATCH] [NPU] fix best practicce docs (#29303) --- .../ascend-npus/best_practice/deepseek_r1.mdx | 4 +++- .../ascend-npus/best_practice/deepseek_v3_2.mdx | 4 +++- .../ascend-npus/best_practice/glm5_1.mdx | 4 +++- .../ascend-npus/best_practice/kimi_k2_6.mdx | 12 +++++++----- .../ascend-npus/best_practice/minimax_m2_5.mdx | 4 +++- .../ascend-npus/best_practice/qwen3-8b.mdx | 4 +++- .../ascend-npus/best_practice/qwen3_235b_a22b.mdx | 5 +++-- .../ascend-npus/best_practice/qwen3_30b_a3b.mdx | 6 ++++-- .../ascend-npus/best_practice/qwen3_32b.mdx | 4 +++- .../ascend-npus/best_practice/qwen3_5_397b.mdx | 4 +++- .../ascend-npus/best_practice/qwen3_6_27b.mdx | 4 +++- .../ascend-npus/best_practice/qwen3_6_35b_a3b.mdx | 4 +++- .../best_practice/qwen3_next_80b_a3b_instruct.mdx | 4 +++- 13 files changed, 44 insertions(+), 19 deletions(-) diff --git a/docs_new/docs/hardware-platforms/ascend-npus/best_practice/deepseek_r1.mdx b/docs_new/docs/hardware-platforms/ascend-npus/best_practice/deepseek_r1.mdx index 403858448..420f26fff 100644 --- a/docs_new/docs/hardware-platforms/ascend-npus/best_practice/deepseek_r1.mdx +++ b/docs_new/docs/hardware-platforms/ascend-npus/best_practice/deepseek_r1.mdx @@ -4,7 +4,9 @@ metatags: description: "Best Practice for DeepSeek-R1 on Ascend NPU" --- -This guide describes the best practice data for DeepSeek-R1 on the Ascend NPU. + +This page focuses on optimal configuration and benchmark results for DeepSeek-R1 on the Ascend NPU. For environment setup, model weight download, feature configuration, and deployment instructions, etc., see the [DeepSeek-R1 Model Tutorial](/docs/hardware-platforms/ascend-npus/model-tutorials/deepseek_r1). + ### Low Latency diff --git a/docs_new/docs/hardware-platforms/ascend-npus/best_practice/deepseek_v3_2.mdx b/docs_new/docs/hardware-platforms/ascend-npus/best_practice/deepseek_v3_2.mdx index dfa316644..6bfb401a4 100644 --- a/docs_new/docs/hardware-platforms/ascend-npus/best_practice/deepseek_v3_2.mdx +++ b/docs_new/docs/hardware-platforms/ascend-npus/best_practice/deepseek_v3_2.mdx @@ -4,7 +4,9 @@ metatags: description: "Best Practice for DeepSeek-V3.2 on Ascend NPU" --- -This guide describes the best practice data for DeepSeek-V3.2 on the Ascend NPU. + +This page focuses on optimal configuration and benchmark results for DeepSeek-V3.2 on the Ascend NPU. For environment setup, model weight download, feature configuration, and deployment instructions, etc., see the [DeepSeek-V3.2 Model Tutorial](/docs/hardware-platforms/ascend-npus/model-tutorials/deepseek_v3_2). + ### Low Latency diff --git a/docs_new/docs/hardware-platforms/ascend-npus/best_practice/glm5_1.mdx b/docs_new/docs/hardware-platforms/ascend-npus/best_practice/glm5_1.mdx index c7850757d..3656098a4 100644 --- a/docs_new/docs/hardware-platforms/ascend-npus/best_practice/glm5_1.mdx +++ b/docs_new/docs/hardware-platforms/ascend-npus/best_practice/glm5_1.mdx @@ -4,7 +4,9 @@ metatags: description: "Best Practice for GLM-5.1 on Ascend NPU" --- -This guide describes the best practice data for GLM-5.1 on the Ascend NPU. + +This page focuses on optimal configuration and benchmark results for GLM-5.1 on the Ascend NPU. For environment setup, model weight download, feature configuration, and deployment instructions, etc., see the [GLM-5.1 Model Tutorial](/docs/hardware-platforms/ascend-npus/model-tutorials/glm_5_1). + ### Low Latency diff --git a/docs_new/docs/hardware-platforms/ascend-npus/best_practice/kimi_k2_6.mdx b/docs_new/docs/hardware-platforms/ascend-npus/best_practice/kimi_k2_6.mdx index 037d169e5..fd7395133 100644 --- a/docs_new/docs/hardware-platforms/ascend-npus/best_practice/kimi_k2_6.mdx +++ b/docs_new/docs/hardware-platforms/ascend-npus/best_practice/kimi_k2_6.mdx @@ -4,7 +4,9 @@ metatags: description: "Best Practice for Kimi-K2.6 on Ascend NPU" --- -This guide describes the best practice data for Kimi-K2.6 on the Ascend NPU. + +This page focuses on optimal configuration and benchmark results for Kimi-K2.6 on the Ascend NPU. For environment setup, model weight download, feature configuration, and deployment instructions, etc., see the [Kimi-K2.6 Model Tutorial](/docs/hardware-platforms/ascend-npus/model-tutorials/kimi_k2_6). + ### Low Latency @@ -1006,7 +1008,7 @@ python3 -m sglang.launch_server \ --host 127.0.0.1 --port 6688 \ --quantization modelslim \ --dtype bfloat16 \ - --model-loader-extra-config {"enable_multithread_load": true} \ + --model-loader-extra-config '{"enable_multithread_load": true}' \ --trust-remote-code \ --device npu \ --attention-backend ascend \ @@ -1116,7 +1118,7 @@ python3 -m sglang.launch_server \ --host 127.0.0.1 --port 6688 \ --quantization modelslim \ --dtype bfloat16 \ - --model-loader-extra-config {"enable_multithread_load": true} \ + --model-loader-extra-config '{"enable_multithread_load": true}' \ --trust-remote-code \ --device npu \ --attention-backend ascend \ @@ -1249,7 +1251,7 @@ python3 -m sglang.launch_server \ --deepep-mode auto \ --cuda-graph-bs 1 2 3 4 5 \ --disable-radix-cache \ - --model-loader-extra-config {"enable_multithread_load": true} \ + --model-loader-extra-config '{"enable_multithread_load": true}' \ --speculative-algorithm EAGLE3 \ --speculative-draft-model-path $DRAFT_MODEL_PATH \ --speculative-num-steps 4 \ @@ -1358,7 +1360,7 @@ python3 -m sglang.launch_server \ --deepep-mode auto \ --cuda-graph-bs 1 2 4 8 12 13 \ --disable-radix-cache \ - --model-loader-extra-config {"enable_multithread_load": true} \ + --model-loader-extra-config '{"enable_multithread_load": true}' \ --speculative-algorithm EAGLE3 \ --speculative-draft-model-path $DRAFT_MODEL_PATH \ --speculative-num-steps 4 \ diff --git a/docs_new/docs/hardware-platforms/ascend-npus/best_practice/minimax_m2_5.mdx b/docs_new/docs/hardware-platforms/ascend-npus/best_practice/minimax_m2_5.mdx index e8a286484..6e9755ae5 100644 --- a/docs_new/docs/hardware-platforms/ascend-npus/best_practice/minimax_m2_5.mdx +++ b/docs_new/docs/hardware-platforms/ascend-npus/best_practice/minimax_m2_5.mdx @@ -4,7 +4,9 @@ metatags: description: "Best Practice for MiniMax-M2.5 on Ascend NPU" --- -This guide describes the best practice data for MiniMax-M2.5 on the Ascend NPU. + +This page focuses on optimal configuration and benchmark results for MiniMax-M2.5 on the Ascend NPU. For environment setup, model weight download, feature configuration, and deployment instructions, etc., see the [MiniMax-M2.5 Model Tutorial](/docs/hardware-platforms/ascend-npus/model-tutorials/minimax_m2_5). + ### Low Latency diff --git a/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3-8b.mdx b/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3-8b.mdx index a73d9685a..952658b20 100644 --- a/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3-8b.mdx +++ b/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3-8b.mdx @@ -4,7 +4,9 @@ metatags: description: "Best Practice for Qwen3-8B on Ascend NPU" --- -This guide describes the best practice data for Qwen3-8B on the Ascend NPU. + +This page focuses on optimal configuration and benchmark results for Qwen3-8B on the Ascend NPU. For environment setup, model weight download, feature configuration, and deployment instructions, etc., see the [Qwen3-8B Model Tutorial](/docs/hardware-platforms/ascend-npus/model-tutorials/qwen3-8b). + ### Low Latency diff --git a/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3_235b_a22b.mdx b/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3_235b_a22b.mdx index ea5340a4d..30ff661d0 100644 --- a/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3_235b_a22b.mdx +++ b/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3_235b_a22b.mdx @@ -4,7 +4,9 @@ metatags: description: "Best Practice for Qwen3-235B-A22B on Ascend NPU" --- -This guide describes the best practice data for Qwen3-235B-A22B on the Ascend NPU. + +This page focuses on optimal configuration and benchmark results for Qwen3-235B-A22B on the Ascend NPU. For environment setup, model weight download, feature configuration, and deployment instructions, etc., see the [Qwen3-235B-A22B Model Tutorial](/docs/hardware-platforms/ascend-npus/model-tutorials/qwen3_235b_a22b). + ### Low Latency @@ -192,7 +194,6 @@ python3 -m sglang.launch_server \ --max-prefill-tokens 458880 \ --sampling-backend ascend \ --ep-dispatch-algorithm static \ - --init-expert-location /root/.cache/modelscope/hub/models/hot_map/235B_3_5k_decode.pt \ --disable-radix-cache \ --moe-a2a-backend ascend_fuseep \ --speculative-algorithm EAGLE3 \ diff --git a/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3_30b_a3b.mdx b/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3_30b_a3b.mdx index ff16db1d1..7020ed265 100644 --- a/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3_30b_a3b.mdx +++ b/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3_30b_a3b.mdx @@ -4,7 +4,9 @@ metatags: description: "Best Practice for Qwen3-30B-A3B on Ascend NPU" --- -This guide describes the best practice data for Qwen3-30B-A3B on the Ascend NPU. + +This page focuses on optimal configuration and benchmark results for Qwen3-30B-A3B on the Ascend NPU. For environment setup, model weight download, feature configuration, and deployment instructions, etc., see the [Qwen3-30B-A3B Model Tutorial](/docs/hardware-platforms/ascend-npus/model-tutorials/qwen3_30b_a3b). + ### Low Latency @@ -71,7 +73,7 @@ source /usr/local/Ascend/nnal/atb/set_env.sh export ASCEND_LAUNCH_BLOCKING=0 export DP_ROUND_ROBIN=1 export GLOO_SOCKET_IFNAME= -export HCCL_ALGO=level0:NA;level1:ring +export HCCL_ALGO="level0:NA;level1:ring" export HCCL_SOCKET_IFNAME= export INF_NAN_MODE_FORCE_DISABLE=1 export PYTORCH_NPU_ALLOC_CONF=expandable_segments:False diff --git a/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3_32b.mdx b/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3_32b.mdx index f29f18bed..4be7a162e 100644 --- a/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3_32b.mdx +++ b/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3_32b.mdx @@ -4,7 +4,9 @@ metatags: description: "Best Practice for Qwen3-32B on Ascend NPU" --- -This guide describes the best practice data for Qwen3-32B on the Ascend NPU. + +This page focuses on optimal configuration and benchmark results for Qwen3-32B on the Ascend NPU. For environment setup, model weight download, feature configuration, and deployment instructions, etc., see the [Qwen3-32B Model Tutorial](/docs/hardware-platforms/ascend-npus/model-tutorials/qwen3_32b). + ### Low Latency diff --git a/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3_5_397b.mdx b/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3_5_397b.mdx index 35c1e2423..f027ebae7 100644 --- a/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3_5_397b.mdx +++ b/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3_5_397b.mdx @@ -4,7 +4,9 @@ metatags: description: "Best Practice for Qwen3.5-397B on Ascend NPU" --- -This guide describes the best practice data for Qwen3.5-397B on the Ascend NPU. + +This page focuses on optimal configuration and benchmark results for Qwen3.5-397B on the Ascend NPU. For environment setup, model weight download, feature configuration, and deployment instructions, etc., see the [Qwen3.5-397B Model Tutorial](/docs/hardware-platforms/ascend-npus/model-tutorials/qwen3_5_397b). + ### Low Latency diff --git a/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3_6_27b.mdx b/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3_6_27b.mdx index 0f4d84823..8871b1876 100644 --- a/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3_6_27b.mdx +++ b/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3_6_27b.mdx @@ -4,7 +4,9 @@ metatags: description: "Best Practice for Qwen3.6-27B on Ascend NPU" --- -This guide describes the best practice data for Qwen3.6-27B on the Ascend NPU. + +This page focuses on optimal configuration and benchmark results for Qwen3.6-27B on the Ascend NPU. For environment setup, model weight download, feature configuration, and deployment instructions, etc., see the [Qwen3.6-27B Model Tutorial](/docs/hardware-platforms/ascend-npus/model-tutorials/qwen3_6_27b). + ### High Throughput diff --git a/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3_6_35b_a3b.mdx b/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3_6_35b_a3b.mdx index e26cd951d..4fdeab6ea 100644 --- a/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3_6_35b_a3b.mdx +++ b/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3_6_35b_a3b.mdx @@ -4,7 +4,9 @@ metatags: description: "Best Practice for Qwen3.6-35B-A3B on Ascend NPU" --- -This guide describes the best practice data for Qwen3.6-35B-A3B on the Ascend NPU. + +This page focuses on optimal configuration and benchmark results for Qwen3.6-35B-A3B on the Ascend NPU. For environment setup, model weight download, feature configuration, and deployment instructions, etc., see the [Qwen3.6-35B-A3B Model Tutorial](/docs/hardware-platforms/ascend-npus/model-tutorials/qwen3_6_35b_a3b). + ### Low Latency diff --git a/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3_next_80b_a3b_instruct.mdx b/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3_next_80b_a3b_instruct.mdx index f768cccc4..c72bedd38 100644 --- a/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3_next_80b_a3b_instruct.mdx +++ b/docs_new/docs/hardware-platforms/ascend-npus/best_practice/qwen3_next_80b_a3b_instruct.mdx @@ -4,7 +4,9 @@ metatags: description: "Best Practice for Qwen3-Next-80B-A3B-Instruct on Ascend NPU" --- -This guide describes the best practice data for Qwen3-Next-80B-A3B-Instruct on the Ascend NPU. + +This page focuses on optimal configuration and benchmark results for Qwen3-Next-80B-A3B-Instruct on the Ascend NPU. For environment setup, model weight download, feature configuration, and deployment instructions, etc., see the [Qwen3-Next-80B-A3B-Instruct Model Tutorial](/docs/hardware-platforms/ascend-npus/model-tutorials/qwen3_next_80b_a3b_instruct). + ### Low Latency