Update readme & fix generate (#4041)

Author: Jintao-Huang

README.md

@@ -51,7 +51,7 @@ You can contact us and communicate with us by adding our group:
 
 
 ## 📝 Introduction
-🍲 ms-swift is an official framework provided by the ModelScope community for fine-tuning and deploying large language models and multi-modal large models. It currently supports the training (pre-training, fine-tuning, human alignment), inference, evaluation, quantization, and deployment of 450+ large models and 150+ multi-modal large models. These large language models (LLMs) include models such as Qwen2.5, InternLM3, GLM4, Mistral, DeepSeek-R1, Yi1.5, TeleChat2, Baichuan2, and Gemma2. The multi-modal LLMs include models such as Qwen2.5-VL, Qwen2-Audio, Llama3.4, Llava, InternVL2.5, MiniCPM-V-2.6, GLM4v, Xcomposer2.5, Yi-VL, DeepSeek-VL2, Phi3.5-Vision, and GOT-OCR2.
+🍲 ms-swift is an official framework provided by the ModelScope community for fine-tuning and deploying large language models and multi-modal large models. It currently supports the training (pre-training, fine-tuning, human alignment), inference, evaluation, quantization, and deployment of 450+ large models and 150+ multi-modal large models. These large language models (LLMs) include models such as Qwen3, Qwen3-MoE, Qwen2.5, InternLM3, GLM4, Mistral, DeepSeek-R1, Yi1.5, TeleChat2, Baichuan2, and Gemma2. The multi-modal LLMs include models such as Qwen2.5-VL, Qwen2-Audio, Llama3.4, Llava, InternVL2.5, MiniCPM-V-2.6, GLM4v, Xcomposer2.5, Yi-VL, DeepSeek-VL2, Phi3.5-Vision, and GOT-OCR2.
 
 🍔 Additionally, ms-swift incorporates the latest training technologies, including lightweight techniques such as LoRA, QLoRA, Llama-Pro, LongLoRA, GaLore, Q-GaLore, LoRA+, LISA, DoRA, FourierFt, ReFT, UnSloth, and Liger, as well as human alignment training methods like DPO, GRPO, RM, PPO, KTO, CPO, SimPO, and ORPO. ms-swift supports acceleration of inference, evaluation, and deployment modules using vLLM and LMDeploy, and it supports model quantization with technologies like GPTQ, AWQ, and BNB. Furthermore, ms-swift offers a Gradio-based Web UI and a wealth of best practices.
 

README_CN.md

@@ -49,7 +49,7 @@
 <img src="asset/discord_qr.jpg" width="200" height="200">  |  <img src="asset/wechat.png" width="200" height="200">
 
 ## 📝 简介
-🍲 ms-swift是魔搭社区提供的大模型与多模态大模型微调部署框架，现已支持450+大模型与150+多模态大模型的训练（预训练、微调、人类对齐）、推理、评测、量化与部署。其中大模型包括：Qwen2.5、InternLM3、GLM4、Mistral、DeepSeek-R1、Yi1.5、TeleChat2、Baichuan2、Gemma2等模型，多模态大模型包括：Qwen2.5-VL、Qwen2-Audio、Llama4、Llava、InternVL2.5、MiniCPM-V-2.6、GLM4v、Xcomposer2.5、Yi-VL、DeepSeek-VL2、Phi3.5-Vision、GOT-OCR2等模型。
+🍲 ms-swift是魔搭社区提供的大模型与多模态大模型微调部署框架，现已支持450+大模型与150+多模态大模型的训练（预训练、微调、人类对齐）、推理、评测、量化与部署。其中大模型包括：Qwen3、Qwen3-MoE、Qwen2.5、InternLM3、GLM4、Mistral、DeepSeek-R1、Yi1.5、TeleChat2、Baichuan2、Gemma2等模型，多模态大模型包括：Qwen2.5-VL、Qwen2-Audio、Llama4、Llava、InternVL2.5、MiniCPM-V-2.6、GLM4v、Xcomposer2.5、Yi-VL、DeepSeek-VL2、Phi3.5-Vision、GOT-OCR2等模型。
 
 🍔 除此之外，ms-swift汇集了最新的训练技术，包括LoRA、QLoRA、Llama-Pro、LongLoRA、GaLore、Q-GaLore、LoRA+、LISA、DoRA、FourierFt、ReFT、UnSloth、和Liger等轻量化训练技术，以及DPO、GRPO、RM、PPO、KTO、CPO、SimPO、ORPO等人类对齐训练方法。ms-swift支持使用vLLM和LMDeploy对推理、评测和部署模块进行加速，并支持使用GPTQ、AWQ、BNB等技术对大模型进行量化。ms-swift还提供了基于Gradio的Web-UI界面及丰富的最佳实践。
 

swift/llm/template/base.py

@@ -535,7 +535,11 @@ def decode_prm(self, input_ids: torch.Tensor, logits: torch.Tensor) -> Any:
         raise NotImplementedError
 
     def generate(self, model, *args, **kwargs):
-        if 'use_model_defaults' in inspect.signature(model.generate).parameters and 'use_model_defaults' not in kwargs:
+        if isinstance(model, PeftModel):
+            signature = inspect.signature(model.model.generate)
+        else:
+            signature = inspect.signature(model.generate)
+        if 'use_model_defaults' in signature.parameters and 'use_model_defaults' not in kwargs:
             kwargs['use_model_defaults'] = False
         return model.generate(*args, **kwargs)
 

swift/trainers/utils.py

@@ -4,6 +4,7 @@
 from types import FunctionType, MethodType
 from typing import List, Union
 
+from peft import PeftModel
 from torch.nn import Module
 
 from swift.utils import get_logger
@@ -13,7 +14,10 @@
 
 def can_return_loss(model: Module) -> bool:
     """Check if a given model can return loss."""
-    signature = inspect.signature(model.forward)
+    if isinstance(model, PeftModel):
+        signature = inspect.signature(model.model.forward)
+    else:
+        signature = inspect.signature(model.forward)
     for p in signature.parameters:
         if p == 'return_loss' and signature.parameters[p].default is True:
             return True
@@ -23,7 +27,10 @@ def can_return_loss(model: Module) -> bool:
 def find_labels(model: Module) -> List[str]:
     """Find the labels used by a given model."""
     model_name = model.__class__.__name__
-    signature = inspect.signature(model.forward)
+    if isinstance(model, PeftModel):
+        signature = inspect.signature(model.model.forward)
+    else:
+        signature = inspect.signature(model.forward)
     if 'QuestionAnswering' in model_name:
         return [p for p in signature.parameters if 'label' in p or p in ('start_positions', 'end_positions')]
     else: