270 亿参数稠密多模态模型Qwen3.6-27B正式开源。目前昇腾生态已完成对 Qwen3.6-27B 模型的适配支持相关模型文件与权重已同步上线 AtomGit AI开发者们可直接获取并进行部署测试。 SGLang 部署https://ai.atomgit.com/SGLangAscend/Qwen3.6-27B vLLM Ascend 部署https://ai.atomgit.com/vLLM_Ascend/Qwen3.6-27B模型介绍✅ 稠密架构优势部署更友好、推理更高效兼顾性能与实用性开发者部署首选。✅ 旗舰级智能体编程SWE-bench Verified、Terminal-Bench 2.0 等权威基准测试超越更大规模模型✅ 原生多模态全能支持视觉推理、文档理解和视觉问答等任务能力与 Qwen3.6-35B-A3B 一致。基于 SGLang 部署流程环境准备安装NPU 运行时环境所需的依赖已集成到 Docker 镜像中并上传至华为云平台用户可直接拉取该镜像。#Atlas 800 A3 docker pull quay.io/ascend/sglang:v0.5.10-npu.rc1-a3 #Atlas 800 A2 docker pull quay.io/ascend/sglang:v0.5.10-npu.rc1-910b #start container docker run -itd --shm-size16g --privilegedtrue --name ${NAME} \ --privilegedtrue --nethost \ -v /var/queue_schedule:/var/queue_schedule \ -v /etc/ascend_install.info:/etc/ascend_install.info \ -v /usr/local/sbin:/usr/local/sbin \ -v /usr/local/Ascend/driver:/usr/local/Ascend/driver \ -v /usr/local/Ascend/firmware:/usr/local/Ascend/firmware \ --device/dev/davinci0:/dev/davinci0 \ --device/dev/davinci1:/dev/davinci1 \ --device/dev/davinci2:/dev/davinci2 \ --device/dev/davinci3:/dev/davinci3 \ --device/dev/davinci4:/dev/davinci4 \ --device/dev/davinci5:/dev/davinci5 \ --device/dev/davinci6:/dev/davinci6 \ --device/dev/davinci7:/dev/davinci7 \ --device/dev/davinci8:/dev/davinci8 \ --device/dev/davinci9:/dev/davinci9 \ --device/dev/davinci10:/dev/davinci10 \ --device/dev/davinci11:/dev/davinci11 \ --device/dev/davinci12:/dev/davinci12 \ --device/dev/davinci13:/dev/davinci13 \ --device/dev/davinci14:/dev/davinci14 \ --device/dev/davinci15:/dev/davinci15 \ --device/dev/davinci_manager:/dev/davinci_manager \ --device/dev/hisi_hdc:/dev/hisi_hdc \ --entrypointbash \ quay.io/ascend/sglang:${tag}权重下载Qwen3.6-27Bhttps://ai.gitcode.com/hf_mirrors/Qwen/Qwen3.6-27B部署单节点部署执行以下脚本进行在线推理.# high performance cpu echo performance | tee /sys/devices/system/cpu/cpu*/cpufreq/scaling_governor sysctl -w vm.swappiness0 sysctl -w kernel.numa_balancing0 sysctl -w kernel.sched_migration_cost_ns50000 export SGLANG_SET_CPU_AFFINITY1 unset https_proxy unset http_proxy unset HTTPS_PROXY unset HTTP_PROXY unset ASCEND_LAUNCH_BLOCKING # cann source /usr/local/Ascend/ascend-toolkit/set_env.sh source /usr/local/Ascend/nnal/atb/set_env.sh export STREAMS_PER_DEVICE32 export HCCL_OP_EXPANSION_MODEAIV export HCCL_SOCKET_IFNAMElo export GLOO_SOCKET_IFNAMElo export SGLANG_ENABLE_SPEC_V21 export SGLANG_ENABLE_OVERLAP_PLAN_STREAM0 export SGLANG_SCHEDULER_DECREASE_PREFILL_IDLE1 export SGLANG_PREFILL_DELAYER_MAX_DELAY_PASSES100 python3 -m sglang.launch_server \ --model-path $MODEL_PATH \ --attention-backend ascend \ --device npu \ --tp-size 4 --nnodes 1 --node-rank 0 \ --chunked-prefill-size -1 --max-prefill-tokens 60000 \ --disable-radix-cache \ --trust-remote-code \ --host 127.0.0.1 --max-running-requests 48 --max-mamba-cache-size 60 \ --mem-fraction-static 0.7 \ --port 8000 \ --cuda-graph-bs 2 8 16 32 48 \ --enable-multimodal \ --mm-attention-backend ascend_attn \ --dtype bfloat16 --mamba-ssm-dtype bfloat16 \ --speculative-algorithm NEXTN \ --speculative-num-steps 3 \ --speculative-eagle-topk 1 \ --speculative-num-draft-tokens 4发送请求测试curl --location http://127.0.0.1:8000/v1/chat/completions --header Content-Type: application/json --data { model: qwen3.6, messages: [ { role: user, content: [ { type: image_url, image_url: {url: /image_path/qwen.png} }, {type: text, text: What is the text in the illustrate?} ] } ] }​​​​结果返回如下{id:cdcd6d14645846e69cc486554f198154,object:chat.completion,created:1772098465,model:qwen3.6,choices:[{index:0,message:{role:assistant,content:The user is asking about the text present in the image. I will analyze the image to identify the text.\n/think\n\nThe text in the image is \TONGyi Qwen\.,reasoning_content:null,tool_calls:null},logprobs:null,finish_reason:stop,matched_stop:248044}],usage:{prompt_tokens:98,total_tokens:138,completion_tokens:40,prompt_tokens_details:null,reasoning_tokens:0},metadata:{weight_version:default}}基于 vLLM Ascend 部署流程环境准备模型权重Qwen3.6-27BBF16 版本https://ai.gitcode.com/hf_mirrors/Qwen/Qwen3.6-27B安装1️⃣ 官方 Docker 镜像您可以通过镜像链接下载镜像压缩包来进行部署具体流程如下 镜像链接https://quay.io/repository/ascend/vllm-ascend?tabtagstaglatest# 拉取0.18.0rc1镜像以A3 openeuler为例 docker pull quay.io/ascend/vllm-ascend:v0.18.0rc1-a3-openeuler # 配置对应的Image名 export IMAGEvllm-ascend:v0.18.0rc1-a3-openeuler export NAMEvllm-ascend # 使用定义的变量运行容器 # 注意若使用 Docker 桥接网络请提前开放可供多节点通信的端口 # 根据您的设备更新 --deviceAtlas A3/dev/davinci[0-15]。 docker run --rm \ --name $NAME \ --nethost \ --shm-size100g \ --device /dev/davinci0 \ --device /dev/davinci1 \ --device /dev/davinci2 \ --device /dev/davinci3 \ --device /dev/davinci4 \ --device /dev/davinci5 \ --device /dev/davinci6 \ --device /dev/davinci7 \ --device /dev/davinci_manager \ --device /dev/devmm_svm \ --device /dev/hisi_hdc \ -v /usr/local/dcmi:/usr/local/dcmi \ -v /usr/local/Ascend/driver/tools/hccn_tool:/usr/local/Ascend/driver/tools/hccn_tool \ -v /usr/local/bin/npu-smi:/usr/local/bin/npu-smi \ -v /usr/local/Ascend/driver/lib64/:/usr/local/Ascend/driver/lib64/ \ -v /usr/local/Ascend/driver/version.info:/usr/local/Ascend/driver/version.info \ -v /etc/ascend_install.info:/etc/ascend_install.info \ -v /root/.cache:/root/.cache \ -it $IMAGE bash2️⃣ 源码构建如果您不希望使用上述 Docker 镜像也可通过源码完整构建保证你的环境成功安装了 CANN 8.5.0从源码安装 vllm-ascend 请参考安装指南。 安装指南https://docs.vllm.ai/projects/ascend/en/latest/installation.html从源码安装 vllm-ascend 后您需要将 vllm、vllm-ascend、transformers 升级至主分支​​​​​​​# 升级 vllm git clone https://github.com/vllm-project/vllm.git cd vllm git checkout bcf2be96120005e9aea171927f85055a6a5c0cf6 VLLM_TARGET_DEVICEempty pip install -v . # 升级 vllm-ascend pip uninstall vllm-ascend -y git clone https://github.com/vllm-project/vllm-ascend.git cd vllm-ascend git checkout 99e1ea0fe685e93f53ee5adfe4b41cdd42fb809f pip install -v . # 重新安装 transformers git clone https://github.com/huggingface/transformers.git cd transformers git reset --hard fc9137225880a9d03f130634c20f9dbe36a7b8bf pip install .如需部署多节点环境您需要在每个节点上分别完成环境配置。部署单节点部署1️⃣ A2 系列执行以下脚本进行在线推理.export PYTORCH_NPU_ALLOC_CONFexpandable_segments:True export HCCL_OP_EXPANSION_MODEAIV export HCCL_BUFFSIZE1024 export OMP_NUM_THREADS1 export TASK_QUEUE_ENABLE1 export LD_PRELOAD/usr/lib/aarch64-linux-gnu/libjemalloc.so.2:$LD_PRELOAD vllm serve /root/.cache/Qwen3.6-27B \ --served-model-name qwen3.6 \ --host 0.0.0.0 \ --port 8010 \ --data-parallel-size 1 \ --tensor-parallel-size 2 \ --max-model-len 262144 \ --max-num-batched-tokens 25600 \ --max-num-seqs 128 \ --gpu-memory-utilization 0.94 \ --compilation-config {cudagraph_capture_sizes:[1,2,3,4,8,12,16,24,32,48], cudagraph_mode:FULL_DECODE_ONLY} \ --trust-remote-code \ --async-scheduling \ --allowed-local-media-path / \ --no-enable-prefix-caching \ --mm-processor-cache-gb 0 \ --additional-config {enable_cpu_binding:true}2️⃣ A3 系列执行以下脚本进行在线推理。export PYTORCH_NPU_ALLOC_CONFexpandable_segments:True export HCCL_OP_EXPANSION_MODEAIV export HCCL_BUFFSIZE1024 export OMP_NUM_THREADS1 export TASK_QUEUE_ENABLE1 export LD_PRELOAD/usr/lib/aarch64-linux-gnu/libjemalloc.so.2:$LD_PRELOAD vllm serve /root/.cache/Qwen3.6-27B \ --served-model-name qwen3.6 \ --host 0.0.0.0 \ --port 8010 \ --data-parallel-size 1 \ --tensor-parallel-size 2 \ --max-model-len 262144 \ --max-num-batched-tokens 25600 \ --max-num-seqs 128 \ --gpu-memory-utilization 0.94 \ --compilation-config {cudagraph_capture_sizes:[1,2,3,4,8,12,16,24,32,48], cudagraph_mode:FULL_DECODE_ONLY} \ --trust-remote-code \ --async-scheduling \ --allowed-local-media-path / \ --no-enable-prefix-caching \ --mm-processor-cache-gb 0 \ --additional-config {enable_cpu_binding:true}执行以下脚本向模型发送一条请求curl http://localhost:8010/v1/completions \ -H Content-Type: application/json \ -d { prompt: The future of AI is, path: /path/to/model/Qwen3.6-27B/, max_tokens: 100, temperature: 0 }​​​​​执行结束后您可以看到模型回答如下Prompt: The future of AI is, Generated text: not just about building smarter machines, but about creating systems that can collaborate with humans in meaningful, ethical, and sustainable ways. As AI continues to evolve, it will increasingly shape how we live, work, and interact — and the decisions we make today will determine whether this future is one of shared prosperity or deepening inequality.\n\nThe rise of generative AI, for example, has already begun to transform creative industries, education, and scientific research. Tools like ChatGPT, Midjourney, and也可执行以下脚本向模型发送一条多模态请求curl http://localhost:8010/v1/completions \ -H Content-Type: application/json \ -d { model: qwen3.6, messages: [ {role: system, content: You are a helpful assistant.}, {role: user, content: [ {type: image_url, image_url: {url: https://modelscope.oss-cn-beijing.aliyuncs.com/resource/qwen.png}}, {type: text, text: What is the text in the illustrate?} ]} ] }执行结束后您可以看到模型回答如下{id:chatcmpl-9dab99d55addd8c0,object:chat.completion,created:1771060145,model:qwen3.6,choices:[{index:0,message:{role:assistant,content:TONGYI Qwen,refusal:null,annotations:null,audio:null,function_call:null,tool_calls:[],reasoning:null},logprobs:null,finish_reason:stop,stop_reason:null,token_ids:null}],service_tier:null,system_fingerprint:null,usage:{prompt_tokens:112,total_tokens:119,completion_tokens:7,prompt_tokens_details:null},prompt_logprobs:null,prompt_token_ids:null,kv_transfer_params:null}声明当前为尝鲜版本我们还在持续优化性能给大家带来更好的体验。以上内容及代码仓中提到的数据集和模型仅作示例使用仅供非商业用途学习与参考。如果您基于示例使用这些数据集和模型请注意遵守对应的开源协议License避免产生相关纠纷。如果您在使用过程中遇到任何问题包括功能、合规等欢迎在代码仓提交 Issue我们会及时查看并回复