diff --git a/llmengine/gemma3_it.py b/llmengine/gemma3_it.py index 1028b29..ac5613f 100644 --- a/llmengine/gemma3_it.py +++ b/llmengine/gemma3_it.py @@ -21,7 +21,7 @@ class Gemma3LLM(MMChatLLM): self.messages = [] self.model_id = model_id -llm_register("google/gemma-3", Gemma3LLM) +llm_register("gemma-3", Gemma3LLM) if __name__ == '__main__': gemma3 = Gemma3LLM('/share/models/google/gemma-3-4b-it') diff --git a/llmengine/server.py b/llmengine/server.py index 568f5ba..0220090 100644 --- a/llmengine/server.py +++ b/llmengine/server.py @@ -2,8 +2,11 @@ from traceback import format_exc import os import sys import argparse -from llmengine.base_embedding import get_llm_class -from llmengine.qwen3embedding import Qwen3Embedding + +from base_chat_llm import BaseChatLLM, get_llm_class +from gemma3_it import Gemma3LLM +from medgemma3_it import MedgemmaLLM +from qwen3 import Qwen3LLM from appPublic.registerfunction import RegisterFunction from appPublic.log import debug, exception @@ -17,7 +20,7 @@ def init(): rf = RegisterFunction() rf.register('chat_completions', chat_completions) -async def embedding(request, params_kw, *params, **kw): +async def chat_completions(request, params_kw, *params, **kw): async def gor(): se = ServerEnv() engine = se.chat_engine