求助；硅基流动key不经过中转可以直接在openwebui中使用吗？

开发调优
,
1

image
image690×221 11 KB

没用new api中转，直接使用的硅基流动的key
openwebui页面报错，但是硅基官网后台有账单显示使用
节点就选的香港，新加坡，台湾，退出vpn后openwenbui还是报错
image
image690×341 43.4 KB

2 Likes
2

image
image1046×517 21.1 KB

1 Like
3

硅基流动没墙啊直接连官网，服务器在北京。我 NextChat 是这样填：https://api.siliconflow.cn/v1/chat/completions#

2 Likes
4

要在openwebui中使用硅基的嵌入向量模型，测试其他硅基模型，报错

5

不是有个函数设置吗,你在自定义一个就可以了 套用openai的函数就可以了

1 Like
6

小白，不懂怎么操作，函数我一次都没用过

7

直接用啊，和openai格式一样的

1 Like
8

这里,官方哪里也有

去get 到你部署的url上就好了

但是那个文档模型向量化,我没搞懂,今天才发现这个api有嵌入api

1 Like
9

但是会报错，硅基后台又能看到使用记录

10 
import asyncio
from typing import Optional
from pydantic import BaseModel, Field


async def emit(emitter, msg, done):
    await emitter(
        {
            "type": "status",
            "data": {
                "done": done,
                "description": msg,
            },
        }
    )


class Filter:
    class Valves(BaseModel):
        priority: int = Field(
            default=0,
            description="Priority level for the filter operations.",
        )
        api_url: str = Field(
            default="https://api.siliconflow.cn/v1",
            description="Base URL for the Siliconflow API.",
        )
        api_key: str = Field(
            default="",
            description="API Key for the Siliconflow API.",
        )

    class UserValves(BaseModel):
        size: str = Field(
            default="1024x1024",
            description="1024x1024, 512x1024, 768x512, 768x1024, 1024x576, 576x1024.",
        )
        steps: int = Field(
            default=35,
            description="The number of inference steps to be performed (1-50).",
        )
        model: str = Field(
            default="black-forest-labs/FLUX.1-dev",
            description="The name of the model. (Pro/black-forest-labs/FLUX.1-schnell, black-forest-labs/FLUX.1-schnell, black-forest-labs/FLUX.1-dev, stabilityai/stable-diffusion-3-medium, stabilityai/stable-diffusion-3-5-large)",
        )
        pnum: int = Field(
            default=3,
            description="The number of pictures.",
        )
        seed: Optional[int] = Field(
            default=None,
            description="The seed.",
        )

    def __init__(self):
        self.valves = self.Valves()

    async def inlet(self, body, __user__, __event_emitter__):
        await emit(__event_emitter__, "Generating prompt, please wait...", False)
        return body

    async def request(self, prompt, __user__):
        url = f"{self.valves.api_url}/image/generations"

        headers = {
            "accept": "application/json",
            "content-type": "application/json",
            "authorization": f"Bearer {self.valves.api_key}",
        }

        payload = {
            "prompt": prompt,
            "model": __user__["valves"].model,
            "image_size": __user__["valves"].size,
            "num_inference_steps": __user__["valves"].steps,
        }

        if seed := __user__["valves"].seed:
            payload["seed"] = seed

        pnum = __user__["valves"].pnum

        async with aiohttp.ClientSession() as sess:
            tasks = [sess.post(url, json=payload, headers=headers) for _ in range(pnum)]
            res = await asyncio.gather(*tasks)
            ret = []

            for i, r in enumerate(res):
                if (s := r.status) == 200:
                    json = await r.json()
                    url = json["images"][0]["url"]
                    ret.append(f"![image{i}]({url})")
                else:
                    text = await r.text()
                    ret.append(f"> The {i} request failed ({s}): {text}.")

            return ret

    async def outlet(self, body, __user__, __event_emitter__):
        await emit(__event_emitter__, f"Generating picture(s), please wait...", False)
        last = body["messages"][-1]
        res = await self.request(last["content"], __user__)

        for r in res:
            last["content"] += f"\n\n{r}"

        await emit(
            __event_emitter__, f"Generated successfully, click to preview!", True
        )

        return body

修改好了,不过这个模型真的有点多啊

1 Like
11

硅基怎么用生图呢，对话可以，生图就提示连不上，

12

整个下午都在赶稿，所以 new api 你连接上了？我 one api 里新建时有个 SiliconFlow（硅基流动官方英文名）然后就跟普通的一样填。

生图的话点我头像那个牛皮癣广告贴就是。。看着图在插件那里配置一下就行。（但我用的是 NextChat，你用 OpenWebUI 的话得搜 OI 的贴那个我没用过）

13

ok，谢谢大佬，搞上了，

14

我的可以直接用