feat: Volcanic Engine Image Model

apps/application/flow/step_node/image_understand_step_node/impl/base_image_understand_node.py

@@ -12,6 +12,7 @@ from application.flow.i_step_node import NodeResult, INode
 from application.flow.step_node.image_understand_step_node.i_image_understand_node import IImageUnderstandNode
 from dataset.models import File
 from setting.models_provider.tools import get_model_instance_by_model_user_id
+from imghdr import what
 
 
 def _write_context(node_variable: Dict, workflow_variable: Dict, node: INode, workflow, answer: str):
@@ -59,8 +60,9 @@ def write_context(node_variable: Dict, workflow_variable: Dict, node: INode, wor
 
 def file_id_to_base64(file_id: str):
     file = QuerySet(File).filter(id=file_id).first()
-    base64_image = base64.b64encode(file.get_byte()).decode("utf-8")
+    file_bytes = file.get_byte()
-    return base64_image
+    base64_image = base64.b64encode(file_bytes).decode("utf-8")
+    return [base64_image, what(None, file_bytes.tobytes())]
 
 
 class BaseImageUnderstandNode(IImageUnderstandNode):
@@ -77,7 +79,7 @@ class BaseImageUnderstandNode(IImageUnderstandNode):
         # 处理不正确的参数
         if image is None or not isinstance(image, list):
             image = []
-
+        print(model_params_setting)
         image_model = get_model_instance_by_model_user_id(model_id, self.flow_params_serializer.data.get('user_id'), **model_params_setting)
         # 执行详情中的历史消息不需要图片内容
         history_message = self.get_history_message_for_details(history_chat_record, dialogue_number)
@@ -152,7 +154,7 @@ class BaseImageUnderstandNode(IImageUnderstandNode):
                 return HumanMessage(
                     content=[
                         {'type': 'text', 'text': data['question']},
-                        *[{'type': 'image_url', 'image_url': {'url': f'data:image/jpeg;base64,{base64_image}'}} for
+                        *[{'type': 'image_url', 'image_url': {'url': f'data:image/{base64_image[1]};base64,{base64_image[0]}'}} for
                           base64_image in image_base64_list]
                     ])
         return HumanMessage(content=chat_record.problem_text)
@@ -167,8 +169,10 @@ class BaseImageUnderstandNode(IImageUnderstandNode):
             for img in image:
                 file_id = img['file_id']
                 file = QuerySet(File).filter(id=file_id).first()
-                base64_image = base64.b64encode(file.get_byte()).decode("utf-8")
+                image_bytes = file.get_byte()
-                images.append({'type': 'image_url', 'image_url': {'url': f'data:image/jpeg;base64,{base64_image}'}})
+                base64_image = base64.b64encode(image_bytes).decode("utf-8")
+                image_format = what(None, image_bytes.tobytes())
+                images.append({'type': 'image_url', 'image_url': {'url': f'data:image/{image_format};base64,{base64_image}'}})
             messages = [HumanMessage(
                 content=[
                     {'type': 'text', 'text': self.workflow_manage.generate_prompt(prompt)},

apps/setting/models_provider/impl/volcanic_engine_model_provider/credential/image.py

@@ -0,0 +1,63 @@
+# coding=utf-8
+import base64
+import os
+from typing import Dict
+
+from langchain_core.messages import HumanMessage
+
+from common import forms
+from common.exception.app_exception import AppApiException
+from common.forms import BaseForm, TooltipLabel
+from setting.models_provider.base_model_provider import BaseModelCredential, ValidCode
+
+class VolcanicEngineImageModelParams(BaseForm):
+    temperature = forms.SliderField(TooltipLabel('温度', '较高的数值会使输出更加随机，而较低的数值会使其更加集中和确定'),
+                                    required=True, default_value=0.95,
+                                    _min=0.1,
+                                    _max=1.0,
+                                    _step=0.01,
+                                    precision=2)
+
+    max_tokens = forms.SliderField(
+        TooltipLabel('输出最大Tokens', '指定模型可生成的最大token个数'),
+        required=True, default_value=1024,
+        _min=1,
+        _max=100000,
+        _step=1,
+        precision=0)
+
+class VolcanicEngineImageModelCredential(BaseForm, BaseModelCredential):
+    api_key = forms.PasswordInputField('API Key', required=True)
+    api_base = forms.TextInputField('API 域名', required=True)
+
+    def is_valid(self, model_type: str, model_name, model_credential: Dict[str, object], provider,
+                 raise_exception=False):
+        model_type_list = provider.get_model_type_list()
+        if not any(list(filter(lambda mt: mt.get('value') == model_type, model_type_list))):
+            raise AppApiException(ValidCode.valid_error.value, f'{model_type} 模型类型不支持')
+
+        for key in ['api_key', 'api_base']:
+            if key not in model_credential:
+                if raise_exception:
+                    raise AppApiException(ValidCode.valid_error.value, f'{key} 字段为必填字段')
+                else:
+                    return False
+        try:
+            model = provider.get_model(model_type, model_name, model_credential)
+            res = model.stream([HumanMessage(content=[{"type": "text", "text": "你好"}])])
+            for chunk in res:
+                print(chunk)
+        except Exception as e:
+            if isinstance(e, AppApiException):
+                raise e
+            if raise_exception:
+                raise AppApiException(ValidCode.valid_error.value, f'校验失败,请检查参数是否正确: {str(e)}')
+            else:
+                return False
+        return True
+
+    def encryption_dict(self, model: Dict[str, object]):
+        return {**model, 'api_key': super().encryption(model.get('api_key', ''))}
+
+    def get_model_params_setting_form(self, model_name):
+        return VolcanicEngineImageModelParams()

apps/setting/models_provider/impl/volcanic_engine_model_provider/credential/tti.py

@@ -0,0 +1,62 @@
+# coding=utf-8
+
+from typing import Dict
+
+from common import forms
+from common.exception.app_exception import AppApiException
+from common.forms import BaseForm, TooltipLabel
+from setting.models_provider.base_model_provider import BaseModelCredential, ValidCode
+
+
+class VolcanicEngineTTIModelGeneralParams(BaseForm):
+    size = forms.SingleSelect(
+        TooltipLabel('图片尺寸',
+                     '宽、高与512差距过大，则出图效果不佳、延迟过长概率显著增加。超分前建议比例及对应宽高：width*height'),
+        required=True,
+        default_value='512*512',
+        option_list=[
+            {'value': '512*512', 'label': '512*512'},
+            {'value': '512*384', 'label': '512*384'},
+            {'value': '384*512', 'label': '384*512'},
+            {'value': '512*341', 'label': '512*341'},
+            {'value': '341*512', 'label': '341*512'},
+            {'value': '512*288', 'label': '512*288'},
+            {'value': '288*512', 'label': '288*512'},
+        ],
+        text_field='label',
+        value_field='value')
+
+
+class VolcanicEngineTTIModelCredential(BaseForm, BaseModelCredential):
+    access_key = forms.PasswordInputField('Access Key', required=True)
+    secret_key = forms.PasswordInputField('Secret Key', required=True)
+
+    def is_valid(self, model_type: str, model_name, model_credential: Dict[str, object], provider,
+                 raise_exception=False):
+        model_type_list = provider.get_model_type_list()
+        if not any(list(filter(lambda mt: mt.get('value') == model_type, model_type_list))):
+            raise AppApiException(ValidCode.valid_error.value, f'{model_type} 模型类型不支持')
+
+        for key in ['access_key', 'secret_key']:
+            if key not in model_credential:
+                if raise_exception:
+                    raise AppApiException(ValidCode.valid_error.value, f'{key} 字段为必填字段')
+                else:
+                    return False
+        try:
+            model = provider.get_model(model_type, model_name, model_credential)
+            model.check_auth()
+        except Exception as e:
+            if isinstance(e, AppApiException):
+                raise e
+            if raise_exception:
+                raise AppApiException(ValidCode.valid_error.value, f'校验失败,请检查参数是否正确: {str(e)}')
+            else:
+                return False
+        return True
+
+    def encryption_dict(self, model: Dict[str, object]):
+        return {**model, 'secret_key': super().encryption(model.get('secret_key', ''))}
+
+    def get_model_params_setting_form(self, model_name):
+        return VolcanicEngineTTIModelGeneralParams()

apps/setting/models_provider/impl/volcanic_engine_model_provider/model/image.py

@@ -0,0 +1,26 @@
+from typing import Dict
+
+from langchain_openai.chat_models import ChatOpenAI
+
+from common.config.tokenizer_manage_config import TokenizerManage
+from setting.models_provider.base_model_provider import MaxKBBaseModel
+
+
+def custom_get_token_ids(text: str):
+    tokenizer = TokenizerManage.get_tokenizer()
+    return tokenizer.encode(text)
+
+
+class VolcanicEngineImage(MaxKBBaseModel, ChatOpenAI):
+
+    @staticmethod
+    def new_instance(model_type, model_name, model_credential: Dict[str, object], **model_kwargs):
+        optional_params = MaxKBBaseModel.filter_optional_params(model_kwargs)
+        return VolcanicEngineImage(
+            model_name=model_name,
+            openai_api_key=model_credential.get('api_key'),
+            openai_api_base=model_credential.get('api_base'),
+            # stream_options={"include_usage": True},
+            streaming=True,
+            **optional_params,
+        )

apps/setting/models_provider/impl/volcanic_engine_model_provider/model/tti.py

@@ -0,0 +1,173 @@
+# coding=utf-8
+
+'''
+requires Python 3.6 or later
+
+pip install asyncio
+pip install websockets
+
+'''
+
+import datetime
+import hashlib
+import hmac
+import json
+import sys
+from typing import Dict
+
+import requests
+from langchain_openai import ChatOpenAI
+
+from setting.models_provider.base_model_provider import MaxKBBaseModel
+from setting.models_provider.impl.base_tti import BaseTextToImage
+
+method = 'POST'
+host = 'visual.volcengineapi.com'
+region = 'cn-north-1'
+endpoint = 'https://visual.volcengineapi.com'
+service = 'cv'
+
+req_key_dict = {
+    'general_v1.4': 'high_aes_general_v14',
+    'general_v2.0': 'high_aes_general_v20',
+    'general_v2.0_L': 'high_aes_general_v20_L',
+    'anime_v1.3': 'high_aes',
+    'anime_v1.3.1': 'high_aes',
+}
+
+
+def sign(key, msg):
+    return hmac.new(key, msg.encode('utf-8'), hashlib.sha256).digest()
+
+
+def getSignatureKey(key, dateStamp, regionName, serviceName):
+    kDate = sign(key.encode('utf-8'), dateStamp)
+    kRegion = sign(kDate, regionName)
+    kService = sign(kRegion, serviceName)
+    kSigning = sign(kService, 'request')
+    return kSigning
+
+
+def formatQuery(parameters):
+    request_parameters_init = ''
+    for key in sorted(parameters):
+        request_parameters_init += key + '=' + parameters[key] + '&'
+    request_parameters = request_parameters_init[:-1]
+    return request_parameters
+
+
+def signV4Request(access_key, secret_key, service, req_query, req_body):
+    if access_key is None or secret_key is None:
+        print('No access key is available.')
+        sys.exit()
+
+    t = datetime.datetime.utcnow()
+    current_date = t.strftime('%Y%m%dT%H%M%SZ')
+    # current_date = '20210818T095729Z'
+    datestamp = t.strftime('%Y%m%d')  # Date w/o time, used in credential scope
+    canonical_uri = '/'
+    canonical_querystring = req_query
+    signed_headers = 'content-type;host;x-content-sha256;x-date'
+    payload_hash = hashlib.sha256(req_body.encode('utf-8')).hexdigest()
+    content_type = 'application/json'
+    canonical_headers = 'content-type:' + content_type + '\n' + 'host:' + host + \
+                        '\n' + 'x-content-sha256:' + payload_hash + \
+                        '\n' + 'x-date:' + current_date + '\n'
+    canonical_request = method + '\n' + canonical_uri + '\n' + canonical_querystring + \
+                        '\n' + canonical_headers + '\n' + signed_headers + '\n' + payload_hash
+    # print(canonical_request)
+    algorithm = 'HMAC-SHA256'
+    credential_scope = datestamp + '/' + region + '/' + service + '/' + 'request'
+    string_to_sign = algorithm + '\n' + current_date + '\n' + credential_scope + '\n' + hashlib.sha256(
+        canonical_request.encode('utf-8')).hexdigest()
+    # print(string_to_sign)
+    signing_key = getSignatureKey(secret_key, datestamp, region, service)
+    # print(signing_key)
+    signature = hmac.new(signing_key, (string_to_sign).encode(
+        'utf-8'), hashlib.sha256).hexdigest()
+    # print(signature)
+
+    authorization_header = algorithm + ' ' + 'Credential=' + access_key + '/' + \
+                           credential_scope + ', ' + 'SignedHeaders=' + \
+                           signed_headers + ', ' + 'Signature=' + signature
+    # print(authorization_header)
+    headers = {'X-Date': current_date,
+               'Authorization': authorization_header,
+               'X-Content-Sha256': payload_hash,
+               'Content-Type': content_type
+               }
+    # print(headers)
+
+    # ************* SEND THE REQUEST *************
+    request_url = endpoint + '?' + canonical_querystring
+
+    print('\nBEGIN REQUEST++++++++++++++++++++++++++++++++++++')
+    print('Request URL = ' + request_url)
+    try:
+        r = requests.post(request_url, headers=headers, data=req_body)
+    except Exception as err:
+        print(f'error occurred: {err}')
+        raise
+    else:
+        print('\nRESPONSE++++++++++++++++++++++++++++++++++++')
+        print(f'Response code: {r.status_code}\n')
+        # 使用 replace 方法将 \u0026 替换为 &
+        resp_str = r.text.replace("\\u0026", "&")
+        if r.status_code != 200:
+            raise Exception(f'Error: {resp_str}')
+        print(f'Response body: {resp_str}\n')
+        return json.loads(resp_str)['data']['image_urls']
+
+
+class VolcanicEngineTextToImage(MaxKBBaseModel, BaseTextToImage):
+    access_key: str
+    secret_key: str
+    model_version: str
+    params: dict
+
+    def __init__(self, **kwargs):
+        super().__init__(**kwargs)
+        self.access_key = kwargs.get('access_key')
+        self.secret_key = kwargs.get('secret_key')
+        self.model_version = kwargs.get('model_version')
+        self.params = kwargs.get('params')
+
+    @staticmethod
+    def new_instance(model_type, model_name, model_credential: Dict[str, object], **model_kwargs):
+        optional_params = {'params': {}}
+        for key, value in model_kwargs.items():
+            if key not in ['model_id', 'use_local', 'streaming']:
+                optional_params['params'][key] = value
+        return VolcanicEngineTextToImage(
+            model_version=model_name,
+            access_key=model_credential.get('access_key'),
+            secret_key=model_credential.get('secret_key'),
+            **optional_params
+        )
+
+    def check_auth(self):
+        res = self.generate_image('生成一张小猫图片')
+        print(res)
+
+    def generate_image(self, prompt: str, negative_prompt: str = None):
+        # 请求Query，按照接口文档中填入即可
+        query_params = {
+            'Action': 'CVProcess',
+            'Version': '2022-08-31',
+        }
+        formatted_query = formatQuery(query_params)
+        size = self.params.pop('size', '512*512').split('*')
+        body_params = {
+            "req_key": req_key_dict[self.model_version],
+            "prompt": prompt,
+            "model_version": self.model_version,
+            "return_url": True,
+            "width": int(size[0]),
+            "height": int(size[1]),
+            **self.params
+        }
+        formatted_body = json.dumps(body_params)
+        return signV4Request(self.access_key, self.secret_key, service, formatted_query, formatted_body)
+
+    def is_cache_model(self):
+        return False

apps/setting/models_provider/impl/volcanic_engine_model_provider/volcanic_engine_model_provider.py

@@ -14,10 +14,15 @@ from setting.models_provider.base_model_provider import IModelProvider, ModelPro
 from setting.models_provider.impl.openai_model_provider.credential.embedding import OpenAIEmbeddingCredential
 from setting.models_provider.impl.openai_model_provider.credential.llm import OpenAILLMModelCredential
 from setting.models_provider.impl.openai_model_provider.model.embedding import OpenAIEmbeddingModel
+from setting.models_provider.impl.volcanic_engine_model_provider.credential.image import \
+    VolcanicEngineImageModelCredential
+from setting.models_provider.impl.volcanic_engine_model_provider.credential.tti import VolcanicEngineTTIModelCredential
 from setting.models_provider.impl.volcanic_engine_model_provider.credential.tts import VolcanicEngineTTSModelCredential
+from setting.models_provider.impl.volcanic_engine_model_provider.model.image import VolcanicEngineImage
 from setting.models_provider.impl.volcanic_engine_model_provider.model.llm import VolcanicEngineChatModel
 from setting.models_provider.impl.volcanic_engine_model_provider.credential.stt import VolcanicEngineSTTModelCredential
 from setting.models_provider.impl.volcanic_engine_model_provider.model.stt import VolcanicEngineSpeechToText
+from setting.models_provider.impl.volcanic_engine_model_provider.model.tti import VolcanicEngineTextToImage
 from setting.models_provider.impl.volcanic_engine_model_provider.model.tts import VolcanicEngineTextToSpeech
 
 from smartdoc.conf import PROJECT_DIR
@@ -25,6 +30,8 @@ from smartdoc.conf import PROJECT_DIR
 volcanic_engine_llm_model_credential = OpenAILLMModelCredential()
 volcanic_engine_stt_model_credential = VolcanicEngineSTTModelCredential()
 volcanic_engine_tts_model_credential = VolcanicEngineTTSModelCredential()
+volcanic_engine_image_model_credential = VolcanicEngineImageModelCredential()
+volcanic_engine_tti_model_credential = VolcanicEngineTTIModelCredential()
 
 model_info_list = [
     ModelInfo('ep-xxxxxxxxxx-yyyy',
@@ -32,6 +39,11 @@ model_info_list = [
               ModelTypeConst.LLM,
               volcanic_engine_llm_model_credential, VolcanicEngineChatModel
               ),
+    ModelInfo('ep-xxxxxxxxxx-yyyy',
+              '用户前往火山方舟的模型推理页面创建推理接入点，这里需要输入ep-xxxxxxxxxx-yyyy进行调用',
+              ModelTypeConst.IMAGE,
+              volcanic_engine_image_model_credential, VolcanicEngineImage
+              ),
     ModelInfo('asr',
               '',
               ModelTypeConst.STT,
@@ -42,6 +54,31 @@ model_info_list = [
               ModelTypeConst.TTS,
               volcanic_engine_tts_model_credential, VolcanicEngineTextToSpeech
               ),
+    ModelInfo('general_v2.0',
+              '通用2.0-文生图',
+              ModelTypeConst.TTI,
+              volcanic_engine_tti_model_credential, VolcanicEngineTextToImage
+              ),
+    ModelInfo('general_v2.0_L',
+              '通用2.0Pro-文生图',
+              ModelTypeConst.TTI,
+              volcanic_engine_tti_model_credential, VolcanicEngineTextToImage
+              ),
+    ModelInfo('general_v1.4',
+              '通用1.4-文生图',
+              ModelTypeConst.TTI,
+              volcanic_engine_tti_model_credential, VolcanicEngineTextToImage
+              ),
+    ModelInfo('anime_v1.3',
+              '动漫1.3.0-文生图',
+              ModelTypeConst.TTI,
+              volcanic_engine_tti_model_credential, VolcanicEngineTextToImage
+              ),
+    ModelInfo('anime_v1.3.1',
+              '动漫1.3.1-文生图',
+              ModelTypeConst.TTI,
+              volcanic_engine_tti_model_credential, VolcanicEngineTextToImage
+              ),
 ]
 
 open_ai_embedding_credential = OpenAIEmbeddingCredential()
@@ -51,8 +88,13 @@ model_info_embedding_list = [
               ModelTypeConst.EMBEDDING, open_ai_embedding_credential,
               OpenAIEmbeddingModel)]
 
-model_info_manage = ModelInfoManage.builder().append_model_info_list(model_info_list).append_default_model_info(
+model_info_manage = (
-    model_info_list[0]).build()
+    ModelInfoManage.builder()
+    .append_model_info_list(model_info_list)
+    .append_default_model_info(model_info_list[0])
+    .append_default_model_info(model_info_list[1])
+    .build()
+)
 
 
 class VolcanicEngineModelProvider(IModelProvider):

ui/src/workflow/nodes/image-understand/index.vue

@@ -25,6 +25,15 @@
               <div>
                 <span>图片理解模型<span class="danger">*</span></span>
               </div>
+              <el-button
+                :disabled="!form_data.model_id"
+                type="primary"
+                link
+                @click="openAIParamSettingDialog(form_data.model_id)"
+                @refreshForm="refreshParam"
+              >
+                {{ $t('views.application.applicationForm.form.paramSetting') }}
+              </el-button>
             </div>
           </template>
           <el-select
@@ -183,6 +192,7 @@
         </el-form-item>
       </el-form>
     </el-card>
+    <AIModeParamSettingDialog ref="AIModeParamSettingDialogRef" @refresh="refreshParam" />
   </NodeContainer>
 </template>
 
@@ -197,6 +207,7 @@ import { app } from '@/main'
 import useStore from '@/stores'
 import NodeCascader from '@/workflow/common/NodeCascader.vue'
 import type { FormInstance } from 'element-plus'
+import AIModeParamSettingDialog from '@/views/application/component/AIModeParamSettingDialog.vue'
 
 const { model } = useStore()
 
@@ -207,6 +218,7 @@ const {
 const props = defineProps<{ nodeModel: any }>()
 const modelOptions = ref<any>(null)
 const providerOptions = ref<Array<Provider>>([])
+const AIModeParamSettingDialogRef = ref<InstanceType<typeof AIModeParamSettingDialog>>()
 
 const aiChatNodeFormRef = ref<FormInstance>()
 const validate = () => {
@@ -281,6 +293,16 @@ function submitDialog(val: string) {
   set(props.nodeModel.properties.node_data, 'prompt', val)
 }
 
+const openAIParamSettingDialog = (modelId: string) => {
+  if (modelId) {
+    AIModeParamSettingDialogRef.value?.open(modelId, id, form_data.value.model_params_setting)
+  }
+}
+
+function refreshParam(data: any) {
+  set(props.nodeModel.properties.node_data, 'model_params_setting', data)
+}
+
 onMounted(() => {
   getModel()
   getProvider()