92 lines
3.2 KiB
Python
Executable File
92 lines
3.2 KiB
Python
Executable File
"""Kimi Provider - 使用阿里云百炼平台"""
|
|
import asyncio
|
|
from typing import AsyncIterator, Iterator
|
|
|
|
from openai import OpenAI
|
|
|
|
from ..base import BaseLLMProvider, LLMMessage, LLMResponse
|
|
|
|
|
|
class KimiProvider(BaseLLMProvider):
|
|
"""Kimi模型Provider - 通过阿里云百炼Coding Plan平台调用
|
|
|
|
支持的模型:
|
|
- kimi-k2.5: 最强模型，适合复杂分析推理
|
|
- kimi-k2.5-latest: 最新版本
|
|
- kimi-latest: 最新轻量版
|
|
"""
|
|
provider_name = "kimi"
|
|
# 阿里云百炼Coding接口
|
|
CODING_BASE_URL = "https://coding.dashscope.aliyuncs.com/v1"
|
|
|
|
# 百炼平台模型名称映射
|
|
MODEL_MAPPING = {
|
|
"kimi-k2.5": "kimi-k2.5",
|
|
"kimi-k2.5-latest": "kimi-k2.5",
|
|
"kimi-latest": "kimi-k2.5",
|
|
"moonshot-v1-8k": "kimi-k2.5",
|
|
"moonshot-v1-128k": "kimi-k2.5",
|
|
}
|
|
|
|
def __init__(
|
|
self,
|
|
api_key: str,
|
|
model: str = "kimi-k2.5",
|
|
temperature: float = 0.7,
|
|
max_tokens: int = 4096,
|
|
api_base: str = None,
|
|
):
|
|
# 映射模型名称
|
|
coding_model = self.MODEL_MAPPING.get(model, model)
|
|
|
|
# 使用传入的base_url或默认的Coding接口
|
|
base_url = api_base or self.CODING_BASE_URL
|
|
|
|
super().__init__(
|
|
api_key=api_key,
|
|
api_base=base_url,
|
|
model=coding_model,
|
|
temperature=temperature,
|
|
max_tokens=max_tokens,
|
|
)
|
|
self.client = OpenAI(api_key=api_key, base_url=base_url)
|
|
print(f"[DEBUG] KimiProvider initialized. api_key_configured={bool(api_key)}, base_url: {base_url}")
|
|
|
|
def call(self, messages: list[LLMMessage], **kwargs) -> LLMResponse:
|
|
response = self.client.chat.completions.create(
|
|
model=self.model,
|
|
messages=self._build_messages(messages),
|
|
temperature=kwargs.get("temperature", self.temperature),
|
|
max_tokens=kwargs.get("max_tokens", self.max_tokens),
|
|
)
|
|
return LLMResponse(
|
|
content=response.choices[0].message.content,
|
|
model=self.model,
|
|
provider=self.provider_name,
|
|
usage={
|
|
"prompt_tokens": response.usage.prompt_tokens,
|
|
"completion_tokens": response.usage.completion_tokens,
|
|
"total_tokens": response.usage.total_tokens,
|
|
},
|
|
finish_reason=response.choices[0].finish_reason,
|
|
)
|
|
|
|
async def acall(self, messages: list[LLMMessage], **kwargs) -> LLMResponse:
|
|
return await asyncio.to_thread(self.call, messages, **kwargs)
|
|
|
|
def stream(self, messages: list[LLMMessage], **kwargs) -> Iterator[str]:
|
|
response = self.client.chat.completions.create(
|
|
model=self.model,
|
|
messages=self._build_messages(messages),
|
|
temperature=kwargs.get("temperature", self.temperature),
|
|
max_tokens=kwargs.get("max_tokens", self.max_tokens),
|
|
stream=True,
|
|
)
|
|
for chunk in response:
|
|
if chunk.choices[0].delta.content:
|
|
yield chunk.choices[0].delta.content
|
|
|
|
async def astream(self, messages: list[LLMMessage], **kwargs) -> AsyncIterator[str]:
|
|
for content in self.stream(messages, **kwargs):
|
|
yield content
|