mirror of
https://github.com/langgenius/dify.git
synced 2024-11-16 11:42:29 +08:00
Add novita.ai as model provider (#4961)
This commit is contained in:
Jason
GitHub
parent
c28d709d7f
commit
b7ff765d8d
|
|
@ -0,0 +1,19 @@
|
|||
<svg width="162" height="36" viewBox="0 0 162 36" fill="none" xmlns="http://www.w3.org/2000/svg">
|
||||
<path fill-rule="evenodd" clip-rule="evenodd" d="M2 0C0.895431 0 0 0.895432 0 2V29.1891C0 30.2937 0.895433 31.1891 2 31.1891H5.51171L16.0608 35.1377C16.7145 35.3824 17.4114 34.8991 17.4114 34.2012V11.3669C17.4114 10.533 16.894 9.78665 16.1131 9.49405L5.51171 5.52152H25.58V31.1891H29.0917C30.1963 31.1891 31.0917 30.2937 31.0917 29.1891V2C31.0917 0.895431 30.1963 0 29.0917 0H2ZM14.6022 23.7351C15.0558 23.956 15.4239 23.6812 15.4239 23.1185C15.4239 22.5557 15.0558 21.9204 14.6022 21.6995C14.1486 21.4775 13.7804 21.7545 13.7804 22.3161C13.7804 22.8777 14.1486 23.513 14.6022 23.7351Z" fill="white"/>
|
||||
<path fill-rule="evenodd" clip-rule="evenodd" d="M2 0C0.895431 0 0 0.895432 0 2V29.1891C0 30.2937 0.895433 31.1891 2 31.1891H5.51171L16.0608 35.1377C16.7145 35.3824 17.4114 34.8991 17.4114 34.2012V11.3669C17.4114 10.533 16.894 9.78665 16.1131 9.49405L5.51171 5.52152H25.58V31.1891H29.0917C30.1963 31.1891 31.0917 30.2937 31.0917 29.1891V2C31.0917 0.895431 30.1963 0 29.0917 0H2ZM14.6022 23.7351C15.0558 23.956 15.4239 23.6812 15.4239 23.1185C15.4239 22.5557 15.0558 21.9204 14.6022 21.6995C14.1486 21.4775 13.7804 21.7545 13.7804 22.3161C13.7804 22.8777 14.1486 23.513 14.6022 23.7351Z" fill="url(#paint0_linear_1473_71)"/>
|
||||
<path d="M55.9397 27.8804H59.0566V19.0803C59.0566 14.9105 56.381 12.7172 52.8228 12.7172C51.0023 12.7172 49.3197 13.4483 48.2991 14.6668V12.9609H45.1546V27.8804H48.2991V19.5406C48.2991 16.8059 49.8162 15.3978 52.1332 15.3978C54.4226 15.3978 55.9397 16.8059 55.9397 19.5406V27.8804Z" fill="#11101A"/>
|
||||
<path fill-rule="evenodd" clip-rule="evenodd" d="M69.7881 12.7172C74.1187 12.7172 77.539 15.7228 77.539 20.4071C77.539 25.0915 74.0083 28.1241 69.6502 28.1241C65.3196 28.1241 62.0372 25.0915 62.0372 20.4071C62.0372 15.7228 65.4575 12.7172 69.7881 12.7172ZM69.7342 15.3979C67.362 15.3979 65.2381 17.0225 65.2381 20.4071C65.2381 23.7918 67.2793 25.4435 69.6514 25.4435C71.996 25.4435 74.313 23.7918 74.313 20.4071C74.313 17.0225 72.0788 15.3979 69.7342 15.3979Z" fill="#11101A"/>
|
||||
<path d="M78.861 12.9609L84.6259 27.8804H88.3772L94.1697 12.9609H90.8321L86.5291 25.1185L82.2261 12.9609H78.861Z" fill="#11101A"/>
|
||||
<path fill-rule="evenodd" clip-rule="evenodd" d="M100.13 9.00761C100.13 10.1178 99.2477 10.9842 98.1443 10.9842C97.0134 10.9842 96.1308 10.1178 96.1308 9.00761C96.1308 7.89745 97.0134 7.03098 98.1443 7.03098C99.2477 7.03098 100.13 7.89745 100.13 9.00761ZM99.6882 27.8804H96.5437V12.9609H99.6882V27.8804Z" fill="#11101A"/>
|
||||
<path d="M104.322 23.7376C104.322 26.7702 106.004 27.8804 108.708 27.8804H111.19V25.308H109.259C107.935 25.308 107.494 24.8477 107.494 23.7376V15.479H111.19V12.9609H107.494V9.25128H104.322V12.9609H102.529V15.479H104.322V23.7376Z" fill="#11101A"/>
|
||||
<path fill-rule="evenodd" clip-rule="evenodd" d="M120.154 28.1241C116.209 28.1241 113.037 24.9561 113.037 20.353C113.037 15.7498 116.209 12.7172 120.209 12.7172C122.774 12.7172 124.539 13.9086 125.477 15.1271V12.9609H128.649V27.8804H125.477V25.6601C124.512 26.9327 122.691 28.1241 120.154 28.1241ZM120.87 25.4435C123.242 25.4435 125.476 23.6293 125.476 20.4071C125.476 17.212 123.242 15.3979 120.87 15.3979C118.526 15.3979 116.264 17.1308 116.264 20.353C116.264 23.5752 118.526 25.4435 120.87 25.4435Z" fill="#11101A"/>
|
||||
<path d="M136.043 26.0933C136.043 24.9832 135.16 24.1167 134.057 24.1167C132.926 24.1167 132.043 24.9832 132.043 26.0933C132.043 27.2035 132.926 28.07 134.057 28.07C135.16 28.07 136.043 27.2035 136.043 26.0933Z" fill="#11101A"/>
|
||||
<path fill-rule="evenodd" clip-rule="evenodd" d="M145.502 28.1241C141.558 28.1241 138.386 24.9561 138.386 20.353C138.386 15.7498 141.558 12.7172 145.557 12.7172C148.123 12.7172 149.888 13.9086 150.826 15.1271V12.9609H153.998V27.8804H150.826V25.6601C149.86 26.9327 148.04 28.1241 145.502 28.1241ZM146.219 25.4435C148.591 25.4435 150.825 23.6293 150.825 20.4071C150.825 17.212 148.591 15.3979 146.219 15.3979C143.874 15.3979 141.612 17.1308 141.612 20.353C141.612 23.5752 143.874 25.4435 146.219 25.4435Z" fill="#11101A"/>
|
||||
<path fill-rule="evenodd" clip-rule="evenodd" d="M161.722 9.00761C161.722 10.1178 160.84 10.9842 159.736 10.9842C158.605 10.9842 157.723 10.1178 157.723 9.00761C157.723 7.89745 158.605 7.03098 159.736 7.03098C160.84 7.03098 161.722 7.89745 161.722 9.00761ZM161.28 27.8804H158.136V12.9609H161.28V27.8804Z" fill="#11101A"/>
|
||||
<defs>
|
||||
<linearGradient id="paint0_linear_1473_71" x1="31" y1="-2" x2="0.975591" y2="14.2625" gradientUnits="userSpaceOnUse">
|
||||
<stop stop-color="#2622FF"/>
|
||||
<stop offset="1" stop-color="#A717FF"/>
|
||||
</linearGradient>
|
||||
</defs>
|
||||
</svg>
|
||||
|
After Width: | Height: | Size: 4.5 KiB |
|
|
@ -0,0 +1,10 @@
|
|||
<svg width="32" height="36" viewBox="0 0 32 36" fill="none" xmlns="http://www.w3.org/2000/svg">
|
||||
<path fill-rule="evenodd" clip-rule="evenodd" d="M2 0C0.895431 0 0 0.895432 0 2V29.1891C0 30.2937 0.895433 31.1891 2 31.1891H5.51171L16.0608 35.1377C16.7145 35.3824 17.4114 34.8991 17.4114 34.2012V11.3669C17.4114 10.533 16.894 9.78665 16.1131 9.49405L5.51171 5.52152H25.58V31.1891H29.0917C30.1963 31.1891 31.0917 30.2937 31.0917 29.1891V2C31.0917 0.895431 30.1963 0 29.0917 0H2ZM14.6022 23.7351C15.0558 23.956 15.4239 23.6812 15.4239 23.1185C15.4239 22.5557 15.0558 21.9204 14.6022 21.6995C14.1486 21.4775 13.7804 21.7545 13.7804 22.3161C13.7804 22.8777 14.1486 23.513 14.6022 23.7351Z" fill="white"/>
|
||||
<path fill-rule="evenodd" clip-rule="evenodd" d="M2 0C0.895431 0 0 0.895432 0 2V29.1891C0 30.2937 0.895433 31.1891 2 31.1891H5.51171L16.0608 35.1377C16.7145 35.3824 17.4114 34.8991 17.4114 34.2012V11.3669C17.4114 10.533 16.894 9.78665 16.1131 9.49405L5.51171 5.52152H25.58V31.1891H29.0917C30.1963 31.1891 31.0917 30.2937 31.0917 29.1891V2C31.0917 0.895431 30.1963 0 29.0917 0H2ZM14.6022 23.7351C15.0558 23.956 15.4239 23.6812 15.4239 23.1185C15.4239 22.5557 15.0558 21.9204 14.6022 21.6995C14.1486 21.4775 13.7804 21.7545 13.7804 22.3161C13.7804 22.8777 14.1486 23.513 14.6022 23.7351Z" fill="url(#paint0_linear_1473_97)"/>
|
||||
<defs>
|
||||
<linearGradient id="paint0_linear_1473_97" x1="31" y1="-2" x2="0.975591" y2="14.2625" gradientUnits="userSpaceOnUse">
|
||||
<stop stop-color="#2622FF"/>
|
||||
<stop offset="1" stop-color="#A717FF"/>
|
||||
</linearGradient>
|
||||
</defs>
|
||||
</svg>
|
||||
|
After Width: | Height: | Size: 1.5 KiB |
|
|
@ -0,0 +1,36 @@
|
|||
model: Nous-Hermes-2-Mixtral-8x7B-DPO
|
||||
label:
|
||||
zh_Hans: Nous-Hermes-2-Mixtral-8x7B-DPO
|
||||
en_US: Nous-Hermes-2-Mixtral-8x7B-DPO
|
||||
model_type: llm
|
||||
features:
|
||||
- agent-thought
|
||||
model_properties:
|
||||
mode: chat
|
||||
context_size: 32768
|
||||
parameter_rules:
|
||||
- name: temperature
|
||||
use_template: temperature
|
||||
min: 0
|
||||
max: 2
|
||||
default: 1
|
||||
- name: top_p
|
||||
use_template: top_p
|
||||
min: 0
|
||||
max: 1
|
||||
default: 1
|
||||
- name: max_tokens
|
||||
use_template: max_tokens
|
||||
min: 1
|
||||
max: 2048
|
||||
default: 512
|
||||
- name: frequency_penalty
|
||||
use_template: frequency_penalty
|
||||
min: -2
|
||||
max: 2
|
||||
default: 0
|
||||
- name: presence_penalty
|
||||
use_template: presence_penalty
|
||||
min: -2
|
||||
max: 2
|
||||
default: 0
|
||||
|
|
@ -0,0 +1,36 @@
|
|||
model: meta-llama/llama-3-70b-instruct
|
||||
label:
|
||||
zh_Hans: meta-llama/llama-3-70b-instruct
|
||||
en_US: meta-llama/llama-3-70b-instruct
|
||||
model_type: llm
|
||||
features:
|
||||
- agent-thought
|
||||
model_properties:
|
||||
mode: chat
|
||||
context_size: 8192
|
||||
parameter_rules:
|
||||
- name: temperature
|
||||
use_template: temperature
|
||||
min: 0
|
||||
max: 2
|
||||
default: 1
|
||||
- name: top_p
|
||||
use_template: top_p
|
||||
min: 0
|
||||
max: 1
|
||||
default: 1
|
||||
- name: max_tokens
|
||||
use_template: max_tokens
|
||||
min: 1
|
||||
max: 2048
|
||||
default: 512
|
||||
- name: frequency_penalty
|
||||
use_template: frequency_penalty
|
||||
min: -2
|
||||
max: 2
|
||||
default: 0
|
||||
- name: presence_penalty
|
||||
use_template: presence_penalty
|
||||
min: -2
|
||||
max: 2
|
||||
default: 0
|
||||
|
|
@ -0,0 +1,36 @@
|
|||
model: meta-llama/llama-3-8b-instruct
|
||||
label:
|
||||
zh_Hans: meta-llama/llama-3-8b-instruct
|
||||
en_US: meta-llama/llama-3-8b-instruct
|
||||
model_type: llm
|
||||
features:
|
||||
- agent-thought
|
||||
model_properties:
|
||||
mode: chat
|
||||
context_size: 8192
|
||||
parameter_rules:
|
||||
- name: temperature
|
||||
use_template: temperature
|
||||
min: 0
|
||||
max: 2
|
||||
default: 1
|
||||
- name: top_p
|
||||
use_template: top_p
|
||||
min: 0
|
||||
max: 1
|
||||
default: 1
|
||||
- name: max_tokens
|
||||
use_template: max_tokens
|
||||
min: 1
|
||||
max: 2048
|
||||
default: 512
|
||||
- name: frequency_penalty
|
||||
use_template: frequency_penalty
|
||||
min: -2
|
||||
max: 2
|
||||
default: 0
|
||||
- name: presence_penalty
|
||||
use_template: presence_penalty
|
||||
min: -2
|
||||
max: 2
|
||||
default: 0
|
||||
48
api/core/model_runtime/model_providers/novita/llm/llm.py
Normal file
48
api/core/model_runtime/model_providers/novita/llm/llm.py
Normal file
|
|
@ -0,0 +1,48 @@
|
|||
from collections.abc import Generator
|
||||
from typing import Optional, Union
|
||||
|
||||
from core.model_runtime.entities.llm_entities import LLMResult
|
||||
from core.model_runtime.entities.message_entities import PromptMessage, PromptMessageTool
|
||||
from core.model_runtime.entities.model_entities import AIModelEntity
|
||||
from core.model_runtime.model_providers.openai_api_compatible.llm.llm import OAIAPICompatLargeLanguageModel
|
||||
|
||||
|
||||
class NovitaLargeLanguageModel(OAIAPICompatLargeLanguageModel):
|
||||
|
||||
def _update_endpoint_url(self, credentials: dict):
|
||||
credentials['endpoint_url'] = "https://api.novita.ai/v3/openai"
|
||||
credentials['extra_headers'] = { 'X-Novita-Source': 'dify.ai' }
|
||||
return credentials
|
||||
|
||||
def _invoke(self, model: str, credentials: dict,
|
||||
prompt_messages: list[PromptMessage], model_parameters: dict,
|
||||
tools: Optional[list[PromptMessageTool]] = None, stop: Optional[list[str]] = None,
|
||||
stream: bool = True, user: Optional[str] = None) \
|
||||
-> Union[LLMResult, Generator]:
|
||||
cred_with_endpoint = self._update_endpoint_url(credentials=credentials)
|
||||
return super()._invoke(model, cred_with_endpoint, prompt_messages, model_parameters, tools, stop, stream, user)
|
||||
def validate_credentials(self, model: str, credentials: dict) -> None:
|
||||
cred_with_endpoint = self._update_endpoint_url(credentials=credentials)
|
||||
self._add_custom_parameters(credentials, model)
|
||||
return super().validate_credentials(model, cred_with_endpoint)
|
||||
|
||||
@classmethod
|
||||
def _add_custom_parameters(cls, credentials: dict, model: str) -> None:
|
||||
credentials['mode'] = 'chat'
|
||||
|
||||
def _generate(self, model: str, credentials: dict, prompt_messages: list[PromptMessage], model_parameters: dict,
|
||||
tools: Optional[list[PromptMessageTool]] = None, stop: Optional[list[str]] = None,
|
||||
stream: bool = True, user: Optional[str] = None) -> Union[LLMResult, Generator]:
|
||||
cred_with_endpoint = self._update_endpoint_url(credentials=credentials)
|
||||
return super()._generate(model, cred_with_endpoint, prompt_messages, model_parameters, tools, stop, stream, user)
|
||||
|
||||
def get_customizable_model_schema(self, model: str, credentials: dict) -> AIModelEntity:
|
||||
cred_with_endpoint = self._update_endpoint_url(credentials=credentials)
|
||||
|
||||
return super().get_customizable_model_schema(model, cred_with_endpoint)
|
||||
|
||||
def get_num_tokens(self, model: str, credentials: dict, prompt_messages: list[PromptMessage],
|
||||
tools: Optional[list[PromptMessageTool]] = None) -> int:
|
||||
cred_with_endpoint = self._update_endpoint_url(credentials=credentials)
|
||||
|
||||
return super().get_num_tokens(model, cred_with_endpoint, prompt_messages, tools)
|
||||
|
|
@ -0,0 +1,36 @@
|
|||
model: lzlv_70b
|
||||
label:
|
||||
zh_Hans: lzlv_70b
|
||||
en_US: lzlv_70b
|
||||
model_type: llm
|
||||
features:
|
||||
- agent-thought
|
||||
model_properties:
|
||||
mode: chat
|
||||
context_size: 4096
|
||||
parameter_rules:
|
||||
- name: temperature
|
||||
use_template: temperature
|
||||
min: 0
|
||||
max: 2
|
||||
default: 1
|
||||
- name: top_p
|
||||
use_template: top_p
|
||||
min: 0
|
||||
max: 1
|
||||
default: 1
|
||||
- name: max_tokens
|
||||
use_template: max_tokens
|
||||
min: 1
|
||||
max: 2048
|
||||
default: 512
|
||||
- name: frequency_penalty
|
||||
use_template: frequency_penalty
|
||||
min: -2
|
||||
max: 2
|
||||
default: 0
|
||||
- name: presence_penalty
|
||||
use_template: presence_penalty
|
||||
min: -2
|
||||
max: 2
|
||||
default: 0
|
||||
|
|
@ -0,0 +1,36 @@
|
|||
model: gryphe/mythomax-l2-13b
|
||||
label:
|
||||
zh_Hans: gryphe/mythomax-l2-13b
|
||||
en_US: gryphe/mythomax-l2-13b
|
||||
model_type: llm
|
||||
features:
|
||||
- agent-thought
|
||||
model_properties:
|
||||
mode: chat
|
||||
context_size: 4096
|
||||
parameter_rules:
|
||||
- name: temperature
|
||||
use_template: temperature
|
||||
min: 0
|
||||
max: 2
|
||||
default: 1
|
||||
- name: top_p
|
||||
use_template: top_p
|
||||
min: 0
|
||||
max: 1
|
||||
default: 1
|
||||
- name: max_tokens
|
||||
use_template: max_tokens
|
||||
min: 1
|
||||
max: 2048
|
||||
default: 512
|
||||
- name: frequency_penalty
|
||||
use_template: frequency_penalty
|
||||
min: -2
|
||||
max: 2
|
||||
default: 0
|
||||
- name: presence_penalty
|
||||
use_template: presence_penalty
|
||||
min: -2
|
||||
max: 2
|
||||
default: 0
|
||||
|
|
@ -0,0 +1,36 @@
|
|||
model: nousresearch/nous-hermes-llama2-13b
|
||||
label:
|
||||
zh_Hans: nousresearch/nous-hermes-llama2-13b
|
||||
en_US: nousresearch/nous-hermes-llama2-13b
|
||||
model_type: llm
|
||||
features:
|
||||
- agent-thought
|
||||
model_properties:
|
||||
mode: chat
|
||||
context_size: 4096
|
||||
parameter_rules:
|
||||
- name: temperature
|
||||
use_template: temperature
|
||||
min: 0
|
||||
max: 2
|
||||
default: 1
|
||||
- name: top_p
|
||||
use_template: top_p
|
||||
min: 0
|
||||
max: 1
|
||||
default: 1
|
||||
- name: max_tokens
|
||||
use_template: max_tokens
|
||||
min: 1
|
||||
max: 2048
|
||||
default: 512
|
||||
- name: frequency_penalty
|
||||
use_template: frequency_penalty
|
||||
min: -2
|
||||
max: 2
|
||||
default: 0
|
||||
- name: presence_penalty
|
||||
use_template: presence_penalty
|
||||
min: -2
|
||||
max: 2
|
||||
default: 0
|
||||
|
|
@ -0,0 +1,36 @@
|
|||
model: teknium/openhermes-2.5-mistral-7b
|
||||
label:
|
||||
zh_Hans: teknium/openhermes-2.5-mistral-7b
|
||||
en_US: teknium/openhermes-2.5-mistral-7b
|
||||
model_type: llm
|
||||
features:
|
||||
- agent-thought
|
||||
model_properties:
|
||||
mode: chat
|
||||
context_size: 4096
|
||||
parameter_rules:
|
||||
- name: temperature
|
||||
use_template: temperature
|
||||
min: 0
|
||||
max: 2
|
||||
default: 1
|
||||
- name: top_p
|
||||
use_template: top_p
|
||||
min: 0
|
||||
max: 1
|
||||
default: 1
|
||||
- name: max_tokens
|
||||
use_template: max_tokens
|
||||
min: 1
|
||||
max: 2048
|
||||
default: 512
|
||||
- name: frequency_penalty
|
||||
use_template: frequency_penalty
|
||||
min: -2
|
||||
max: 2
|
||||
default: 0
|
||||
- name: presence_penalty
|
||||
use_template: presence_penalty
|
||||
min: -2
|
||||
max: 2
|
||||
default: 0
|
||||
|
|
@ -0,0 +1,36 @@
|
|||
model: microsoft/wizardlm-2-8x22b
|
||||
label:
|
||||
zh_Hans: microsoft/wizardlm-2-8x22b
|
||||
en_US: microsoft/wizardlm-2-8x22b
|
||||
model_type: llm
|
||||
features:
|
||||
- agent-thought
|
||||
model_properties:
|
||||
mode: chat
|
||||
context_size: 65535
|
||||
parameter_rules:
|
||||
- name: temperature
|
||||
use_template: temperature
|
||||
min: 0
|
||||
max: 2
|
||||
default: 1
|
||||
- name: top_p
|
||||
use_template: top_p
|
||||
min: 0
|
||||
max: 1
|
||||
default: 1
|
||||
- name: max_tokens
|
||||
use_template: max_tokens
|
||||
min: 1
|
||||
max: 2048
|
||||
default: 512
|
||||
- name: frequency_penalty
|
||||
use_template: frequency_penalty
|
||||
min: -2
|
||||
max: 2
|
||||
default: 0
|
||||
- name: presence_penalty
|
||||
use_template: presence_penalty
|
||||
min: -2
|
||||
max: 2
|
||||
default: 0
|
||||
31
api/core/model_runtime/model_providers/novita/novita.py
Normal file
31
api/core/model_runtime/model_providers/novita/novita.py
Normal file
|
|
@ -0,0 +1,31 @@
|
|||
import logging
|
||||
|
||||
from core.model_runtime.entities.model_entities import ModelType
|
||||
from core.model_runtime.errors.validate import CredentialsValidateFailedError
|
||||
from core.model_runtime.model_providers.__base.model_provider import ModelProvider
|
||||
|
||||
logger = logging.getLogger(__name__)
|
||||
|
||||
|
||||
class NovitaProvider(ModelProvider):
|
||||
def validate_provider_credentials(self, credentials: dict) -> None:
|
||||
"""
|
||||
Validate provider credentials
|
||||
if validate failed, raise exception
|
||||
|
||||
:param credentials: provider credentials, credentials form defined in `provider_credential_schema`.
|
||||
"""
|
||||
try:
|
||||
model_instance = self.get_model_instance(ModelType.LLM)
|
||||
|
||||
# Use `meta-llama/llama-3-8b-instruct` model for validate,
|
||||
# no matter what model you pass in, text completion model or chat model
|
||||
model_instance.validate_credentials(
|
||||
model='meta-llama/llama-3-8b-instruct',
|
||||
credentials=credentials
|
||||
)
|
||||
except CredentialsValidateFailedError as ex:
|
||||
raise ex
|
||||
except Exception as ex:
|
||||
logger.exception(f'{self.get_provider_schema().provider} credentials validate failed')
|
||||
raise ex
|
||||
28
api/core/model_runtime/model_providers/novita/novita.yaml
Normal file
28
api/core/model_runtime/model_providers/novita/novita.yaml
Normal file
|
|
@ -0,0 +1,28 @@
|
|||
provider: novita
|
||||
label:
|
||||
en_US: novita.ai
|
||||
icon_small:
|
||||
en_US: icon_s_en.svg
|
||||
icon_large:
|
||||
en_US: icon_l_en.svg
|
||||
background: "#eadeff"
|
||||
help:
|
||||
title:
|
||||
en_US: Get your API key from novita.ai
|
||||
zh_Hans: 从 novita.ai 获取 API Key
|
||||
url:
|
||||
en_US: https://novita.ai/dashboard/key?utm_source=dify
|
||||
supported_model_types:
|
||||
- llm
|
||||
configurate_methods:
|
||||
- predefined-model
|
||||
provider_credential_schema:
|
||||
credential_form_schemas:
|
||||
- variable: api_key
|
||||
required: true
|
||||
label:
|
||||
en_US: API Key
|
||||
type: secret-input
|
||||
placeholder:
|
||||
zh_Hans: 在此输入您的 API Key
|
||||
en_US: Enter your API Key
|
||||
|
|
@ -74,7 +74,7 @@ class OAIAPICompatLargeLanguageModel(_CommonOAI_API_Compat, LargeLanguageModel):
|
|||
tools=tools,
|
||||
stop=stop,
|
||||
stream=stream,
|
||||
user=user
|
||||
user=user,
|
||||
)
|
||||
|
||||
def get_num_tokens(self, model: str, credentials: dict, prompt_messages: list[PromptMessage],
|
||||
|
|
@ -280,6 +280,12 @@ class OAIAPICompatLargeLanguageModel(_CommonOAI_API_Compat, LargeLanguageModel):
|
|||
'Content-Type': 'application/json',
|
||||
'Accept-Charset': 'utf-8',
|
||||
}
|
||||
extra_headers = credentials.get('extra_headers')
|
||||
if extra_headers is not None:
|
||||
headers = {
|
||||
**headers,
|
||||
**extra_headers,
|
||||
}
|
||||
|
||||
api_key = credentials.get('api_key')
|
||||
if api_key:
|
||||
|
|
|
|||
123
api/tests/integration_tests/model_runtime/novita/test_llm.py
Normal file
123
api/tests/integration_tests/model_runtime/novita/test_llm.py
Normal file
|
|
@ -0,0 +1,123 @@
|
|||
import os
|
||||
from collections.abc import Generator
|
||||
|
||||
import pytest
|
||||
|
||||
from core.model_runtime.entities.llm_entities import LLMResult, LLMResultChunk, LLMResultChunkDelta
|
||||
from core.model_runtime.entities.message_entities import (
|
||||
AssistantPromptMessage,
|
||||
PromptMessageTool,
|
||||
SystemPromptMessage,
|
||||
UserPromptMessage,
|
||||
)
|
||||
from core.model_runtime.errors.validate import CredentialsValidateFailedError
|
||||
from core.model_runtime.model_providers.novita.llm.llm import NovitaLargeLanguageModel
|
||||
|
||||
|
||||
def test_validate_credentials():
|
||||
model = NovitaLargeLanguageModel()
|
||||
|
||||
with pytest.raises(CredentialsValidateFailedError):
|
||||
model.validate_credentials(
|
||||
model='meta-llama/llama-3-8b-instruct',
|
||||
credentials={
|
||||
'api_key': 'invalid_key',
|
||||
'mode': 'chat'
|
||||
}
|
||||
)
|
||||
|
||||
model.validate_credentials(
|
||||
model='meta-llama/llama-3-8b-instruct',
|
||||
credentials={
|
||||
'api_key': os.environ.get('NOVITA_API_KEY'),
|
||||
'mode': 'chat'
|
||||
}
|
||||
)
|
||||
|
||||
|
||||
def test_invoke_model():
|
||||
model = NovitaLargeLanguageModel()
|
||||
|
||||
response = model.invoke(
|
||||
model='meta-llama/llama-3-8b-instruct',
|
||||
credentials={
|
||||
'api_key': os.environ.get('NOVITA_API_KEY'),
|
||||
'mode': 'completion'
|
||||
},
|
||||
prompt_messages=[
|
||||
SystemPromptMessage(
|
||||
content='You are a helpful AI assistant.',
|
||||
),
|
||||
UserPromptMessage(
|
||||
content='Who are you?'
|
||||
)
|
||||
],
|
||||
model_parameters={
|
||||
'temperature': 1.0,
|
||||
'top_p': 0.5,
|
||||
'max_tokens': 10,
|
||||
},
|
||||
stop=['How'],
|
||||
stream=False,
|
||||
user="novita"
|
||||
)
|
||||
|
||||
assert isinstance(response, LLMResult)
|
||||
assert len(response.message.content) > 0
|
||||
|
||||
|
||||
def test_invoke_stream_model():
|
||||
model = NovitaLargeLanguageModel()
|
||||
|
||||
response = model.invoke(
|
||||
model='meta-llama/llama-3-8b-instruct',
|
||||
credentials={
|
||||
'api_key': os.environ.get('NOVITA_API_KEY'),
|
||||
'mode': 'chat'
|
||||
},
|
||||
prompt_messages=[
|
||||
SystemPromptMessage(
|
||||
content='You are a helpful AI assistant.',
|
||||
),
|
||||
UserPromptMessage(
|
||||
content='Who are you?'
|
||||
)
|
||||
],
|
||||
model_parameters={
|
||||
'temperature': 1.0,
|
||||
'top_k': 2,
|
||||
'top_p': 0.5,
|
||||
'max_tokens': 100
|
||||
},
|
||||
stream=True,
|
||||
user="novita"
|
||||
)
|
||||
|
||||
assert isinstance(response, Generator)
|
||||
|
||||
for chunk in response:
|
||||
assert isinstance(chunk, LLMResultChunk)
|
||||
assert isinstance(chunk.delta, LLMResultChunkDelta)
|
||||
assert isinstance(chunk.delta.message, AssistantPromptMessage)
|
||||
|
||||
|
||||
def test_get_num_tokens():
|
||||
model = NovitaLargeLanguageModel()
|
||||
|
||||
num_tokens = model.get_num_tokens(
|
||||
model='meta-llama/llama-3-8b-instruct',
|
||||
credentials={
|
||||
'api_key': os.environ.get('NOVITA_API_KEY'),
|
||||
},
|
||||
prompt_messages=[
|
||||
SystemPromptMessage(
|
||||
content='You are a helpful AI assistant.',
|
||||
),
|
||||
UserPromptMessage(
|
||||
content='Hello World!'
|
||||
)
|
||||
]
|
||||
)
|
||||
|
||||
assert isinstance(num_tokens, int)
|
||||
assert num_tokens == 21
|
||||
|
|
@ -0,0 +1,21 @@
|
|||
import os
|
||||
|
||||
import pytest
|
||||
|
||||
from core.model_runtime.errors.validate import CredentialsValidateFailedError
|
||||
from core.model_runtime.model_providers.novita.novita import NovitaProvider
|
||||
|
||||
|
||||
def test_validate_provider_credentials():
|
||||
provider = NovitaProvider()
|
||||
|
||||
with pytest.raises(CredentialsValidateFailedError):
|
||||
provider.validate_provider_credentials(
|
||||
credentials={}
|
||||
)
|
||||
|
||||
provider.validate_provider_credentials(
|
||||
credentials={
|
||||
'api_key': os.environ.get('NOVITA_API_KEY'),
|
||||
}
|
||||
)
|
||||
Loading…
Reference in New Issue
Block a user