[misc] update license year & fix llama pro (#6814)

* fix llamapro script

* change year

Former-commit-id: d9ae594178796994d400a5f207d6499712816f89

scripts/api_example/test_image.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

scripts/api_example/test_toolcall.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

scripts/convert_ckpt/llamafy_baichuan2.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -19,15 +19,10 @@ from typing import Any, Dict
 
 import fire
 import torch
+from huggingface_hub import split_torch_state_dict_into_shards
 from safetensors.torch import save_file
 from tqdm import tqdm
-from transformers.modeling_utils import (
-    SAFE_WEIGHTS_INDEX_NAME,
-    SAFE_WEIGHTS_NAME,
-    WEIGHTS_INDEX_NAME,
-    WEIGHTS_NAME,
-    shard_checkpoint,
-)
+from transformers.modeling_utils import SAFE_WEIGHTS_INDEX_NAME, SAFE_WEIGHTS_NAME, WEIGHTS_INDEX_NAME, WEIGHTS_NAME
 
 
 CONFIG_NAME = "config.json"
@@ -40,34 +35,42 @@ def save_weight(input_dir: str, output_dir: str, shard_size: str, save_safetenso
             shard_weight = torch.load(os.path.join(input_dir, filepath), map_location="cpu")
             baichuan2_state_dict.update(shard_weight)
 
-    llama2_state_dict: Dict[str, torch.Tensor] = OrderedDict()
+    llama_state_dict: Dict[str, torch.Tensor] = OrderedDict()
     for key, value in tqdm(baichuan2_state_dict.items(), desc="Convert format"):
         if "W_pack" in key:
             proj_size = value.size(0) // 3
-            llama2_state_dict[key.replace("W_pack", "q_proj")] = value[:proj_size, :]
-            llama2_state_dict[key.replace("W_pack", "k_proj")] = value[proj_size : 2 * proj_size, :]
-            llama2_state_dict[key.replace("W_pack", "v_proj")] = value[2 * proj_size :, :]
+            llama_state_dict[key.replace("W_pack", "q_proj")] = value[:proj_size, :]
+            llama_state_dict[key.replace("W_pack", "k_proj")] = value[proj_size : 2 * proj_size, :]
+            llama_state_dict[key.replace("W_pack", "v_proj")] = value[2 * proj_size :, :]
         elif "lm_head" in key:
-            llama2_state_dict[key] = torch.nn.functional.normalize(value)
+            llama_state_dict[key] = torch.nn.functional.normalize(value)
         else:
-            llama2_state_dict[key] = value
+            llama_state_dict[key] = value
 
     weights_name = SAFE_WEIGHTS_NAME if save_safetensors else WEIGHTS_NAME
-    shards, index = shard_checkpoint(llama2_state_dict, max_shard_size=shard_size, weights_name=weights_name)
-
-    for shard_file, shard in tqdm(shards.items(), desc="Save weights"):
+    filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors")
+    state_dict_split = split_torch_state_dict_into_shards(
+        llama_state_dict, filename_pattern=filename_pattern, max_shard_size=shard_size
+    )
+    for shard_file, tensors in tqdm(state_dict_split.filename_to_tensors.items(), desc="Save weights"):
+        shard = {tensor: llama_state_dict[tensor].contiguous() for tensor in tensors}
         if save_safetensors:
             save_file(shard, os.path.join(output_dir, shard_file), metadata={"format": "pt"})
         else:
             torch.save(shard, os.path.join(output_dir, shard_file))
 
-    if index is None:
-        print(f"Model weights saved in {os.path.join(output_dir, WEIGHTS_NAME)}")
+    if not state_dict_split.is_sharded:
+        print(f"Model weights saved in {os.path.join(output_dir, weights_name)}.")
     else:
+        index = {
+            "metadata": state_dict_split.metadata,
+            "weight_map": state_dict_split.tensor_to_filename,
+        }
         index_name = SAFE_WEIGHTS_INDEX_NAME if save_safetensors else WEIGHTS_INDEX_NAME
         with open(os.path.join(output_dir, index_name), "w", encoding="utf-8") as f:
             json.dump(index, f, indent=2, sort_keys=True)
-        print(f"Model weights saved in {output_dir}")
+
+        print(f"Model weights saved in {output_dir}.")
 
 
 def save_config(input_dir: str, output_dir: str):
@@ -81,6 +84,7 @@ def save_config(input_dir: str, output_dir: str):
 
     with open(os.path.join(output_dir, CONFIG_NAME), "w", encoding="utf-8") as f:
         json.dump(llama2_config_dict, f, indent=2)
+
     print(f"Model config saved in {os.path.join(output_dir, CONFIG_NAME)}")
 
 

scripts/convert_ckpt/llamafy_qwen.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -19,16 +19,11 @@ from typing import Any, Dict
 
 import fire
 import torch
+from huggingface_hub import split_torch_state_dict_into_shards
 from safetensors import safe_open
 from safetensors.torch import save_file
 from tqdm import tqdm
-from transformers.modeling_utils import (
-    SAFE_WEIGHTS_INDEX_NAME,
-    SAFE_WEIGHTS_NAME,
-    WEIGHTS_INDEX_NAME,
-    WEIGHTS_NAME,
-    shard_checkpoint,
-)
+from transformers.modeling_utils import SAFE_WEIGHTS_INDEX_NAME, SAFE_WEIGHTS_NAME, WEIGHTS_INDEX_NAME, WEIGHTS_NAME
 from transformers.utils import check_min_version
 
 
@@ -49,60 +44,68 @@ def save_weight(input_dir: str, output_dir: str, shard_size: str, save_safetenso
                 for key in f.keys():
                     qwen_state_dict[key] = f.get_tensor(key)
 
-    llama2_state_dict: Dict[str, torch.Tensor] = OrderedDict()
+    llama_state_dict: Dict[str, torch.Tensor] = OrderedDict()
     torch_dtype = None
     for key, value in tqdm(qwen_state_dict.items(), desc="Convert format"):
         if torch_dtype is None:
             torch_dtype = value.dtype
         if "wte" in key:
-            llama2_state_dict["model.embed_tokens.weight"] = value
+            llama_state_dict["model.embed_tokens.weight"] = value
         elif "ln_f" in key:
-            llama2_state_dict["model.norm.weight"] = value
+            llama_state_dict["model.norm.weight"] = value
         else:
             key = key.replace("transformer.h", "model.layers")
             if "attn.c_attn" in key:
                 proj_size = value.size(0) // 3
-                llama2_state_dict[key.replace("attn.c_attn", "self_attn.q_proj")] = value[:proj_size, ...]
-                llama2_state_dict[key.replace("attn.c_attn", "self_attn.k_proj")] = value[
+                llama_state_dict[key.replace("attn.c_attn", "self_attn.q_proj")] = value[:proj_size, ...]
+                llama_state_dict[key.replace("attn.c_attn", "self_attn.k_proj")] = value[
                     proj_size : 2 * proj_size, ...
                 ]
-                llama2_state_dict[key.replace("attn.c_attn", "self_attn.v_proj")] = value[2 * proj_size :, ...]
+                llama_state_dict[key.replace("attn.c_attn", "self_attn.v_proj")] = value[2 * proj_size :, ...]
             elif "attn.c_proj" in key:
-                llama2_state_dict[key.replace("attn.c_proj", "self_attn.o_proj")] = value
-                llama2_state_dict[key.replace("attn.c_proj.weight", "self_attn.o_proj.bias")] = torch.zeros_like(
+                llama_state_dict[key.replace("attn.c_proj", "self_attn.o_proj")] = value
+                llama_state_dict[key.replace("attn.c_proj.weight", "self_attn.o_proj.bias")] = torch.zeros_like(
                     value[:, 0]
                 ).squeeze()
             elif "ln_1" in key:
-                llama2_state_dict[key.replace("ln_1", "input_layernorm")] = value
+                llama_state_dict[key.replace("ln_1", "input_layernorm")] = value
             elif "ln_2" in key:
-                llama2_state_dict[key.replace("ln_2", "post_attention_layernorm")] = value
+                llama_state_dict[key.replace("ln_2", "post_attention_layernorm")] = value
             elif "mlp.w1" in key:
-                llama2_state_dict[key.replace("mlp.w1", "mlp.up_proj")] = value
+                llama_state_dict[key.replace("mlp.w1", "mlp.up_proj")] = value
             elif "mlp.w2" in key:
-                llama2_state_dict[key.replace("mlp.w2", "mlp.gate_proj")] = value
+                llama_state_dict[key.replace("mlp.w2", "mlp.gate_proj")] = value
             elif "mlp.c_proj" in key:
-                llama2_state_dict[key.replace("mlp.c_proj", "mlp.down_proj")] = value
+                llama_state_dict[key.replace("mlp.c_proj", "mlp.down_proj")] = value
             elif "lm_head" in key:
-                llama2_state_dict[key] = value
+                llama_state_dict[key] = value
             else:
                 raise KeyError(f"Unable to process key {key}")
 
     weights_name = SAFE_WEIGHTS_NAME if save_safetensors else WEIGHTS_NAME
-    shards, index = shard_checkpoint(llama2_state_dict, max_shard_size=shard_size, weights_name=weights_name)
-
-    for shard_file, shard in tqdm(shards.items(), desc="Save weights"):
+    filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors")
+    state_dict_split = split_torch_state_dict_into_shards(
+        llama_state_dict, filename_pattern=filename_pattern, max_shard_size=shard_size
+    )
+    for shard_file, tensors in tqdm(state_dict_split.filename_to_tensors.items(), desc="Save weights"):
+        shard = {tensor: llama_state_dict[tensor].contiguous() for tensor in tensors}
         if save_safetensors:
             save_file(shard, os.path.join(output_dir, shard_file), metadata={"format": "pt"})
         else:
             torch.save(shard, os.path.join(output_dir, shard_file))
 
-    if index is None:
-        print(f"Model weights saved in {os.path.join(output_dir, weights_name)}")
+    if not state_dict_split.is_sharded:
+        print(f"Model weights saved in {os.path.join(output_dir, weights_name)}.")
     else:
+        index = {
+            "metadata": state_dict_split.metadata,
+            "weight_map": state_dict_split.tensor_to_filename,
+        }
         index_name = SAFE_WEIGHTS_INDEX_NAME if save_safetensors else WEIGHTS_INDEX_NAME
         with open(os.path.join(output_dir, index_name), "w", encoding="utf-8") as f:
             json.dump(index, f, indent=2, sort_keys=True)
-        print(f"Model weights saved in {output_dir}")
+
+        print(f"Model weights saved in {output_dir}.")
 
     return str(torch_dtype).replace("torch.", "")
 
@@ -134,6 +137,7 @@ def save_config(input_dir: str, output_dir: str, torch_dtype: str):
 
     with open(os.path.join(output_dir, CONFIG_NAME), "w", encoding="utf-8") as f:
         json.dump(llama2_config_dict, f, indent=2)
+
     print(f"Model config saved in {os.path.join(output_dir, CONFIG_NAME)}")
 
 

scripts/llama_pro.py

@@ -1,4 +1,4 @@
-# Copyright 2024 Tencent Inc. and the LlamaFactory team.
+# Copyright 2025 Tencent Inc. and the LlamaFactory team.
 #
 # This code is inspired by the Tencent's LLaMA-Pro library.
 # https://github.com/TencentARC/LLaMA-Pro/blob/main/scripts/block_expansion.py
@@ -18,20 +18,15 @@
 import json
 import os
 from collections import OrderedDict
-from typing import TYPE_CHECKING
+from typing import TYPE_CHECKING, Dict
 
 import fire
 import torch
+from huggingface_hub import split_torch_state_dict_into_shards
 from safetensors.torch import save_file
 from tqdm import tqdm
 from transformers import AutoConfig, AutoModelForCausalLM, AutoTokenizer, PreTrainedModel
-from transformers.modeling_utils import (
-    SAFE_WEIGHTS_INDEX_NAME,
-    SAFE_WEIGHTS_NAME,
-    WEIGHTS_INDEX_NAME,
-    WEIGHTS_NAME,
-    shard_checkpoint,
-)
+from transformers.modeling_utils import SAFE_WEIGHTS_INDEX_NAME, SAFE_WEIGHTS_NAME, WEIGHTS_INDEX_NAME, WEIGHTS_NAME
 
 
 if TYPE_CHECKING:
@@ -46,41 +41,36 @@ def block_expansion(
     model_name_or_path: str,
     output_dir: str,
     num_expand: int,
-    shard_size: str = "2GB",
+    shard_size: str = "5GB",
     save_safetensors: bool = True,
 ):
     r"""
-    Performs block expansion for LLaMA, Mistral, Qwen1.5 or Yi models.
+    Performs block expansion for LLaMA, Mistral, Qwen2 or Yi models.
     Usage: python llama_pro.py --model_name_or_path meta-llama/Llama-2-7b-hf --output_dir llama2_pro --num_expand 8
     """
-    config: "PretrainedConfig" = AutoConfig.from_pretrained(model_name_or_path)
+    config: "PretrainedConfig" = AutoConfig.from_pretrained(model_name_or_path, trust_remote_code=True)
     num_layers = getattr(config, "num_hidden_layers")
-    setattr(config, "num_hidden_layers", num_layers + num_expand)
-    config.save_pretrained(output_dir)
-
-    tokenizer = AutoTokenizer.from_pretrained(model_name_or_path)
-    tokenizer.save_pretrained(output_dir)
-
-    config = AutoConfig.from_pretrained(model_name_or_path)  # load the original one
-    if save_safetensors:
-        setattr(config, "tie_word_embeddings", False)  # safetensors does not allow shared weights
-
-    model = AutoModelForCausalLM.from_pretrained(
-        model_name_or_path,
-        config=config,
-        torch_dtype="auto",
-        trust_remote_code=True,
-        low_cpu_mem_usage=True,
-    )
-    assert isinstance(model, PreTrainedModel)  # type hint
-    state_dict = model.state_dict()
-
     if num_layers % num_expand != 0:
         raise ValueError(f"`num_layers` {num_layers} should be divisible by `num_expand` {num_expand}.")
 
+    setattr(config, "num_hidden_layers", num_layers + num_expand)
+    config.save_pretrained(output_dir)
+
+    tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, trust_remote_code=True)
+    tokenizer.save_pretrained(output_dir)
+
+    print(f"Expanding model of {num_layers} layers to {num_layers + num_expand} layers.")
+    model = AutoModelForCausalLM.from_pretrained(
+        model_name_or_path, torch_dtype="auto", device_map="cpu", trust_remote_code=True, low_cpu_mem_usage=True
+    )
+    assert isinstance(model, PreTrainedModel)  # type hint
+    if save_safetensors and getattr(model.config, "tie_word_embeddings", False):
+        del model.lm_head  # safetensors does not allow shared weights
+
     split = num_layers // num_expand
     layer_cnt = 0
-    output_state_dict = OrderedDict()
+    state_dict = model.state_dict()
+    output_state_dict: Dict[str, "torch.Tensor"] = OrderedDict()
     for i in range(num_layers):
         for key, value in state_dict.items():
             if f".{i:d}." in key:
@@ -104,17 +94,24 @@ def block_expansion(
             output_state_dict[key] = value
 
     weights_name = SAFE_WEIGHTS_NAME if save_safetensors else WEIGHTS_NAME
-    shards, index = shard_checkpoint(output_state_dict, max_shard_size=shard_size, weights_name=weights_name)
-
-    for shard_file, shard in tqdm(shards.items(), desc="Save weights"):
+    filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors")
+    state_dict_split = split_torch_state_dict_into_shards(
+        output_state_dict, filename_pattern=filename_pattern, max_shard_size=shard_size
+    )
+    for shard_file, tensors in tqdm(state_dict_split.filename_to_tensors.items(), desc="Save weights"):
+        shard = {tensor: output_state_dict[tensor].contiguous() for tensor in tensors}
         if save_safetensors:
             save_file(shard, os.path.join(output_dir, shard_file), metadata={"format": "pt"})
         else:
             torch.save(shard, os.path.join(output_dir, shard_file))
 
-    if index is None:
+    if not state_dict_split.is_sharded:
         print(f"Model weights saved in {os.path.join(output_dir, weights_name)}.")
     else:
+        index = {
+            "metadata": state_dict_split.metadata,
+            "weight_map": state_dict_split.tensor_to_filename,
+        }
         index_name = SAFE_WEIGHTS_INDEX_NAME if save_safetensors else WEIGHTS_INDEX_NAME
         with open(os.path.join(output_dir, index_name), "w", encoding="utf-8") as f:
             json.dump(index, f, indent=2, sort_keys=True)

scripts/loftq_init.py

@@ -1,4 +1,4 @@
-# Copyright 2024 HuggingFace Inc. and the LlamaFactory team.
+# Copyright 2025 HuggingFace Inc. and the LlamaFactory team.
 #
 # This code is based on the HuggingFace's PEFT library.
 # https://github.com/huggingface/peft/blob/v0.10.0/examples/loftq_finetuning/quantize_save_load.py

scripts/pissa_init.py

@@ -1,4 +1,4 @@
-# Copyright 2024 HuggingFace Inc. and the LlamaFactory team.
+# Copyright 2025 HuggingFace Inc. and the LlamaFactory team.
 #
 # This code is based on the HuggingFace's PEFT library.
 # https://github.com/huggingface/peft/blob/v0.11.0/examples/pissa_finetuning/preprocess.py

scripts/stat_utils/cal_flops.py

@@ -1,4 +1,4 @@
-# Copyright 2024 Microsoft Corporation and the LlamaFactory team.
+# Copyright 2025 Microsoft Corporation and the LlamaFactory team.
 #
 # This code is inspired by the Microsoft's DeepSpeed library.
 # https://www.deepspeed.ai/tutorials/flops-profiler/

scripts/stat_utils/cal_lr.py

@@ -1,4 +1,4 @@
-# Copyright 2024 imoneoi and the LlamaFactory team.
+# Copyright 2025 imoneoi and the LlamaFactory team.
 #
 # This code is inspired by the imoneoi's OpenChat library.
 # https://github.com/imoneoi/openchat/blob/3.6.0/ochat/training_deepspeed/train.py

scripts/stat_utils/cal_mfu.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

scripts/stat_utils/cal_ppl.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

scripts/stat_utils/length_cdf.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

scripts/vllm_infer.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

setup.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/api.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/__init__.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/api/app.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/api/chat.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/api/common.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/api/protocol.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/chat/__init__.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/chat/base_engine.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/chat/hf_engine.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/chat/vllm_engine.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/cli.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/data/__init__.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/data/aligner.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/data/data_utils.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/data/formatter.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/data/loader.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/data/parser.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/data/preprocess.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/data/processors/feedback.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/data/processors/pairwise.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/data/processors/processor_utils.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/data/processors/supervised.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/data/processors/unsupervised.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/data/template.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -642,7 +642,7 @@ _register_template(
 
 _register_template(
     name="empty",
-    efficient_eos=True,
+    format_assistant=StringFormatter(slots=["{{content}}"]),
 )
 
 

src/llamafactory/data/tool_utils.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/eval/evaluator.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # This code is inspired by the Dan's test library.
 # https://github.com/hendrycks/test/blob/master/evaluate_flan.py

src/llamafactory/eval/template.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/extras/constants.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/extras/ploting.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/hparams/__init__.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/hparams/evaluation_args.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/hparams/finetuning_args.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/hparams/generating_args.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/launcher.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/model/__init__.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/model/adapter.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/model/loader.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/model/model_utils/attention.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/model/model_utils/embedding.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/model/model_utils/liger_kernel.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/model/model_utils/misc.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/model/model_utils/mod.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/model/model_utils/moe.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/model/model_utils/unsloth.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/model/model_utils/valuehead.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/model/patcher.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/train/callbacks.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/train/dpo/__init__.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/train/kto/__init__.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/train/ppo/__init__.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/train/ppo/ppo_utils.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/train/pt/__init__.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/train/pt/trainer.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/train/rm/__init__.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/train/rm/metric.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/train/sft/__init__.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/train/test_utils.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/train/tuner.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/webui/chatter.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/webui/common.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/webui/components/__init__.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/webui/components/chatbot.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/webui/components/data.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/webui/components/eval.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/webui/components/export.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/webui/components/infer.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/webui/components/top.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/webui/components/train.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/webui/control.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/webui/css.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/webui/engine.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/webui/interface.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/webui/locales.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/webui/manager.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/llamafactory/webui/runner.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/train.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

src/webui.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

tests/data/processors/test_feedback.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

tests/data/processors/test_pairwise.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

tests/data/processors/test_processor_utils.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

tests/data/processors/test_supervised.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

tests/data/processors/test_unsupervised.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

tests/data/test_collator.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

tests/data/test_formatter.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

tests/data/test_mm_plugin.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

tests/data/test_template.py

@@ -1,4 +1,4 @@
-# Copyright 2024 the LlamaFactory team.
+# Copyright 2025 the LlamaFactory team.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.