关于多模态模型放进dsh壳子无法视图 #1354
Allen-David
started this conversation in
General
Replies: 3 comments 1 reply
|
需要在配置的时候声明大模型支持图片识别 [text, image],具体配置请查看下这里 dsh-llm-pi-ai |
1 reply
|
楼主这个场景现在可以有两条路:如果渠道本身确实支持图片,按上面的官方说明给模型声明 Pi 生态里的 dsh plugin --profile web add pi2dsh@0.11.0
dsh plugin --profile web add @kassing/pi-vision
export VISION_BRIDGE_BASE_URL=https://openrouter.ai/api/v1
export VISION_BRIDGE_MODEL=qwen/qwen2.5-vl-72b-instruct
export VISION_BRIDGE_API_KEY=your-openrouter-key重启 Web 后,模型选择器里会在原 DSH 路由旁出现对应的 这条链路已经在真实 DSH Web 对话里做过贴图闭环验证:图片进入伴生路由、视觉结果完成注入、原文本模型在零图片输入下正确回答颜色。仓库里有可复现示例和截图:Vision Bridge example。 |
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment


Uh oh!
There was an error while loading. Please reload this page.
用的是字节的模型 不管ark-code-latest用auto 还是确定的doubao 多模态模型 glm5.3等都无法识别图片
All reactions