first commit

LICENSE

+MIT License
+
+Copyright (c) [year] [fullname]
+
+Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
\ No newline at end of file

README.md

+# Apriel
+## 论文
+[Apriel](https://arxiv.org/pdf/2510.01141)
+## 模型简介
+Apriel-1.6-15B-Thinker是ServiceNow公司Apriel系列轻量级语言模型中的一款升级版多模态推理模型，基于前代Apriel-1.5-15B-Thinker架构演进。该模型在文本与图像推理能力上实现显著提升，其综合性能可对标规模达自身10倍的大型模型。沿袭前代模型的技术路径，该版本继续受益于跨文本与图像领域的持续大规模预训练。并进一步实施了以监督微调（SFT）和强化学习（RL）为核心的后期训练策略。Apriel-1.6在保持高效推理token使用的前提下实现了前沿性能表现。相较于Apriel-1.5-15B-Thinker，该模型在维持或提升各项任务性能指标的同时，将推理token消耗量降低了30%以上。  
+
+该模型仅凭150亿参数规模，即可实现与远超其体量的大模型相当甚至更优的智能评分。在性能图表中，它稳固地位于最具吸引力的象限，在计算效率与顶级推理能力之间取得卓越平衡。实际应用中，这意味着Apriel-1.6-15B-Thinker能以重量级竞品几分之一的计算与部署成本，提供强劲的性能表现与深度推理能力，使其成为现实场景（尤其是企业级应用）中极具效率优势的选择方案。
+
+
+
+<div align=center>
+    <img src="./doc/perf.png"/>
+</div>
+
+## 环境依赖
+
+| 软件 | 版本 |
+| :------: | :------: |
+| DTK | 25.04.2 |
+| python | 3.10.12 |
+| transformers | >=4.57.1 |
+| vllm |  0.9.2+das.opt1.dtk25042 |
+| torch | 2.5.1+das.opt1.dtk25042 |
+| triton | 3.1+das.opt1.3c5d12d.dtk25041 |
+| flash_attn | 2.6.1+das.opt1.dtk2504 |
+| flash_mla | 1.0.0+das.opt1.dtk25042 |
+
+当前仅支持镜像:
+- 挂载地址`-v`根据实际模型情况修改
+
+```bash
+docker run -it --shm-size 60g --network=host --name Apriel --privileged --device=/dev/kfd --device=/dev/dri --device=/dev/mkfd --group-add video --cap-add=SYS_PTRACE --security-opt seccomp=unconfined -u root -v /opt/hyhal/:/opt/hyhal/:ro -v /path/your_code_path/:/path/your_code_path/  image.sourcefind.cn:5000/dcu/admin/base/vllm:0.9.2-ubuntu22.04-dtk25.04.2-py3.10 bash
+```
+更多镜像可前往[光源](https://sourcefind.cn/#/service-list)下载使用。
+
+关于本项目DCU显卡所需的特殊深度学习库可从[光合](https://developer.sourcefind.cn/tool/)开发者社区下载安装。
+
+## 数据集
+暂无
+
+## 训练
+暂无
+
+## 推理
+
+### vllm
+#### 单机推理
+可参考vllm_serve.sh脚本
+```bash
+## serve启动
+vllm serve ServiceNow-AI/Apriel-1.6-15b-Thinker    --trust-remote-code --dtype bfloat16 --max-seq-len-to-capture 32768 -tp 1   --max-model-len 32768
+## client访问
+可参考vllm_cilent.sh
+curl -X POST "http://localhost:8000/v1/chat/completions"   -H "Content-Type: application/json"     --data '{
+                "model": "ServiceNow-AI/Apriel-1.6-15b-Thinker",
+                "messages": [
+                        {
+                                "role": "user",
+                                "content": [
+                                        {
+                                                "type": "text",
+                                                "text": "Describe this image in one sentence."
+                                        },
+                                        {
+                                                "type": "image_url",
+                                                "image_url": {
+                                                        "url": "https://img-s.msn.cn/tenant/amp/entityid/AA1S6LMz.img?w=640&h=427&m=6"
+                                                }
+                                        }
+                                ]
+                        }
+                ]
+        }'
+```
+
+## 效果展示
+<div align=center>
+    <img src="./doc/result.png"/>
+</div>
+
+### 精度
+DCU与GPU精度一致，推理框架：vllm。
+
+## 预训练权重
+| 模型名称  | 权重大小  | DCU型号  | 最低卡数需求 |下载地址|
+|:-----:|:----------:|:----------:|:---------------------:|:----------:|
+| Apriel-1.6-15b-Thinker | 15B | K100AI | 1 | [下载地址](https://huggingface.co/ServiceNow-AI/Apriel-1.6-15b-Thinker) |
+
+## 源码仓库及问题反馈
+- https://developer.sourcefind.cn/codes/modelzoo/apriel_vllm
+
+## 参考资料
+- https://hf-mirror.com/blog/ServiceNow-AI/apriel-1p6-15b-thinker

model.properties

+# 模型唯一标识
+modelCode=1888
+# 模型名称
+modelName=apriel_vllm
+# 模型描述
+modelDescription=Apriel-1.6-15B-Thinker是ServiceNow公司Apriel系列轻量级语言模型中的一款升级版多模态推理模型，基于前代Apriel-1.5-15B-Thinker架构演进。
+processType=推理
+# 算法类别
+appScenario=多模态
+# 框架类型
+frameType=vllm
+# 加速卡类型
+accelerateType=K100AI

vllm_cilent.sh

+curl -X POST "http://localhost:8000/v1/chat/completions"   -H "Content-Type: application/json"     --data '{
+                "model": "ServiceNow-AI/Apriel-1.6-15b-Thinker",
+                "messages": [
+                        {
+                                "role": "user",
+                                "content": [
+                                        {
+                                                "type": "text",
+                                                "text": "Describe this image in one sentence."
+                                        },
+                                        {
+                                                "type": "image_url",
+                                                "image_url": {
+                                                        "url": "https://img-s.msn.cn/tenant/amp/entityid/AA1S6LMz.img?w=640&h=427&m=6"
+                                                }
+                                        }
+                                ]
+                        }
+                ]
+        }'
\ No newline at end of file

vllm_serve.sh

+vllm serve ServiceNow-AI/Apriel-1.6-15b-Thinker    --trust-remote-code --dtype bfloat16 --max-seq-len-to-capture 32768 -tp 1   --max-model-len 32768
\ No newline at end of file