Ms-swift

中文版

新特性
1. 支持使用`swift app`开启可视化推理创空间，参考[这里](https://github.com/modelscope/ms-swift/tree/main/examples/app)
2. 支持大模型的RM和PPO训练，参考[这里](https://github.com/modelscope/ms-swift/tree/main/examples/train/rlhf)
3. 支持`SequenceClassification`模型（含BERT）的BNB/GPTQ量化，参考[这里](https://github.com/modelscope/ms-swift/tree/main/examples/export/quantize/bert)
4. 支持reward model的推理、部署和BNB/GPTQ量化

新模型
1. ZhipuAI/cogagent-9b-20241220
2. Reward Models: Shanghai_AI_Laboratory/internlm2-1_8b-reward系列, Qwen/Qwen2-Math-RM-72B系列, AI-ModelScope/Skywork-Reward-Llama-3.1-8B系列, AI-ModelScope/GRM_Llama3.1_8B_rewardmodel-ft系列
3. AIDC-AI/Ovis1.6-Gemma2-27B, AIDC-AI/Ovis1.6-Llama3.2-3B
4. PowerInfer/SmallThinker-3B-Preview

新数据集
1. PowerInfer/LONGCOT-Refine-500K, PowerInfer/QWQ-LONGCOT-500K

English Version

New Features

1. Support for using `swift app` to launch a visual inference creative space, see [here](https://github.com/modelscope/ms-swift/tree/main/examples/app)
2. Support for RM and PPO training of large models, see [here](https://github.com/modelscope/ms-swift/tree/main/examples/train/rlhf)
3. Support for BNB/GPTQ quantization of `SequenceClassification` models (including BERT), see [here](https://github.com/modelscope/ms-swift/tree/main/examples/export/quantize/bert)
4. Support for inference, deployment, and BNB/GPTQ quantization of reward models

New Models

1. ZhipuAI/cogagent-9b-20241220
2. Reward Models: Shanghai_AI_Laboratory/internlm2-1_8b-reward series, Qwen/Qwen2-Math-RM-72B series, AI-ModelScope/Skywork-Reward-Llama-3.1-8B series, AI-ModelScope/GRM_Llama3.1_8B_rewardmodel-ft series
3. AIDC-AI/Ovis1.6-Gemma2-27B, AIDC-AI/Ovis1.6-Llama3.2-3B
4. PowerInfer/SmallThinker-3B-Preview

New Datasets

1. PowerInfer/LONGCOT-Refine-500K, PowerInfer/QWQ-LONGCOT-500K

What's Changed
* Fix app-ui dropdown by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2787
* fix multi-lora by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2790
* fix stream infer by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2793
* fix some web-ui bugs by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2794
* support swift app by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2792
* fix pt batch infer by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2800
* fix world_size by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2801
* update base_model deploy example by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2803
* fix glm4v by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2806
* fix swift deploy log error (repeat log) by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2808
* support ZhipuAI/cogagent-9b-20241220 by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2810
* fix citest by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2812
* fix enable_cache by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2813
* update docs (specific model arguments) by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2822
* add 'right' option for 'truncation_strategy' by zsxm1998 in https://github.com/modelscope/ms-swift/pull/2754
* Fix glm4v suffix by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2829
* Update padding side by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2832
* Update base_to_chat shell by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2833
* Fix bugs by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2838
* Fix some bugs by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2848
* support reward_model by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2849
* Move optimizer to create_optimizer by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2851
* fix post_init by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2855
* fix cache_name_file by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2856
* fix telechat template by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2857
* Update more models by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2852
* Support quant bert reward by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2859
* fix jsonl writer by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2860
* support reward model train by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2862
* fix vllm video by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2864
* support mps by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2866
* Update agent demo by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2867
* fix bugs by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2869
* Support ppo by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2783
* update citest by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2873
* fix dataset cache bugs by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2876

New Contributors
* zsxm1998 made their first contribution in https://github.com/modelscope/ms-swift/pull/2754

**Full Changelog**: https://github.com/modelscope/ms-swift/compare/v3.0.1...v3.0.2

中文版

新特性
1. 支持`SequenceClassification`模型的训练、推理和部署。可以查看以下例子：[qwen2.5](https://github.com/modelscope/ms-swift/tree/main/examples/train/seq_cls/qwen2_5)，[bert](https://github.com/modelscope/ms-swift/tree/main/examples/train/seq_cls/bert)。
2. LlamaPro支持多模态模型. 例如：qwen2vl、internvl2.5、llama3-vision等。

新模型
1. Qwen/QVQ-72B-Preview
2. iic/DocOwl2
3. OpenGVLab/InternVL2-Pretrain-Models, OpenGVLab/InternVL2_5-4B-AWQ系列, OpenGVLab/InternVL2_5-1B-MPO系列
4. deepseek-ai/DeepSeek-V3系列
5. answerdotai/ModernBERT-base系列
6. AI-ModelScope/paligemma2-3b-pt-224系列, AI-ModelScope/paligemma2-3b-ft-docci-448系列
7. AI-ModelScope/Skywork-o1-Open-Llama-3.1-8B

English Version

New Features:
1. Support for training, inference, and deployment of `SequenceClassification` models. You can check the following examples: [qwen2.5](https://github.com/modelscope/ms-swift/tree/main/examples/train/seq_cls/qwen2_5), [bert](https://github.com/modelscope/ms-swift/tree/main/examples/train/seq_cls/bert).
2. LlamaPro supports multimodal models, such as qwen2vl, internvl2.5, and llama3-vision.

New Models:
1. Qwen/QVQ-72B-Preview
2. iic/DocOwl2
3. OpenGVLab/InternVL2-Pretrain-Models, OpenGVLab/InternVL2_5-4B-AWQ series, OpenGVLab/InternVL2_5-1B-MPO series
4. deepseek-ai/DeepSeek-V3 series
5. answerdotai/ModernBERT-base series
6. AI-ModelScope/paligemma2-3b-pt-224 series, AI-ModelScope/paligemma2-3b-ft-docci-448 series
7. AI-ModelScope/Skywork-o1-Open-Llama-3.1-8B

What's Changed
* Fix mplug owl2, molmo by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2724
* fix batch_infer pad_token & florence by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2725
* Support qwen agent format by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2722
* Support more internvl2.5 awq/mpo & internvl2 pretrain model by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2726
* support iic/DocOwl2 by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2728
* update examples by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2730
* remove files by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2732
* support paligemma2 by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2735
* fix windows by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2733
* support multi-modal llamapro by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2738
* support AI-ModelScope/Skywork-o1-Open-Llama-3.1-8B by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2739
* Fix windows encoding gbk by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2741
* fix docs multimodal by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2742
* support SequenceClassification & update QVQ-72B-Preview by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2747
* fix web-ui by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2758
* fix bugs by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2761
* fix shell by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2764
* fix app-ui by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2765
* support modern_bert & support bert deploy by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2767
* fix alpaca by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2771
* support txt by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2772
* fix telechat2 template by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2775
* Fix deepspeed by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2778
* fix qwen2vl by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2779
* Fix app ui by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2780
* support deepseek-v3 by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2781
* Fix app-ui by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2784


**Full Changelog**: https://github.com/modelscope/ms-swift/compare/v3.0.0...v3.0.1

中文版

架构修改与新特性：
具体可以查看这里: https://swift.readthedocs.io/zh-cn/latest/Instruction/ReleaseNote3.0.html

新模型：
1. OpenGVLab/InternVL2_5-1B等系列模型
2. LLM-Research/Llama-3.3-70B-Instruct
3. BAAI/Emu3-Gen
4. deepseek-ai/DeepSeek-V2.5-1210, deepseek-ai/deepseek-vl2等系列模型
5. Shanghai_AI_Laboratory/internlm-xcomposer2d5-ol-7b
6. InfiniAI/Megrez-3b-Instruct, InfiniAI/Megrez-3B-Omni
7. TeleAI/TeleChat2-3B等系列模型

English Version

Architecture Modifications and New Features:
For more details, please visit: https://swift.readthedocs.io/en/latest/Instruction/ReleaseNote3.0.html

New Models:
1. OpenGVLab/InternVL2_5-1B series models
2. LLM-Research/Llama-3.3-70B-Instruct
3. BAAI/Emu3-Gen
4. deepseek-ai/DeepSeek-V2.5-1210, deepseek-ai/deepseek-vl2 series models
5. Shanghai_AI_Laboratory/internlm-xcomposer2d5-ol-7b
6. InfiniAI/Megrez-3b-Instruct, InfiniAI/Megrez-3B-Omni
7. TeleAI/TeleChat2-3B series models

What's Changed
* Refactor All Codes and bump version to 3.0 by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2030
* fix doc by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2545
* fix manifest by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2546
* add doc 2.x by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2548
* fix ui by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2549
* fix infer by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2550
* Refactor mllm by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2543
* fix ui by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2552
* Fix ui by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2556
* Update ddp infer doc by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2557
* fix docs by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2558
* Fix docs by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2561
* fix log by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2564
* Fix the command line parameter doc by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2565
* fix context by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2568
* Documents Updates by yrk111222 in https://github.com/modelscope/ms-swift/pull/2574
* Revert "Documents Updates" by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2576
* fix hub param by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2572
* Fix bugs by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2573
* Support internvl2.5 by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2575
* update english docs by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2577
* fix en docs by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2580
* fix docs & add custom example by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2581
* fix custom example by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2582
* support llama3.3 by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2584
* update acc_strategy & fix citest by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2583
* Support peft0.14 by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2587
* update infer/deploy examples by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2588
* add image images mapping by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2594
* update llm sft notebook by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2599
* fix notebook by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2600
* Fix streaming by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2601
* Emu3 gen train by mi804 in https://github.com/modelscope/ms-swift/pull/2602
* compat mllm notebook by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2604
* Temporarily remove torchacc. by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2606
* update docs by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2607
* train and infer scripts for emu3_gen by mi804 in https://github.com/modelscope/ms-swift/pull/2610
* Uodate Document by yrk111222 in https://github.com/modelscope/ms-swift/pull/2615
* update memory usage of emu3-gen by mi804 in https://github.com/modelscope/ms-swift/pull/2611
* move prepare_model by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2614
* Update mllm notebook by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2617
* Support all-embedding / all-norm by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2619
* fix lmdeploy==0.5.* by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2621
* Support deepseek-ai/DeepSeek-V2.5-1210 by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2624
* fix use_reentrant gradient_checkpointing by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2625
* support reward model by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2628
* fix add_default_tag by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2631
* fix dataset by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2636
* fix bugs & update openbuddy models & update docs by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2638
* fix app-ui by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2641
* Fix post encode by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2643
* fix bugs by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2645
* update truncation_strategy by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2647
* fix swift/Infinity-Instruct by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2651
* Support LoRA-GA by lxline in https://github.com/modelscope/ms-swift/pull/2650
* support deepseek_vl2 by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2654
* fix swift/SlimOrca by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2656
* fix swift/SlimOrca by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2657
* support Shanghai_AI_Laboratory/internlm-xcomposer2d5-ol-7b:audio by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2658
* support Shanghai_AI_Laboratory/internlm-xcomposer2d5-ol-7b:base by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2660
* fix hub by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2661
* fix liger by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2666
* support megrez by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2667
* fix unsloth resume training by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2668
* fix dataset by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2670
* Fix bugs by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2671
* fix deepseek_vl2 by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2675
* support adapters by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2633
* Support megrez omni by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2674
* fix docs by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2679
* fix megrez_omni by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2680
* fix infer by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2681
* Fix bugs by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2687
* Update readme by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2579
* update wechat by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2694
* fix readme by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2696
* Fix web-ui by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2693
* Fix readme by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2697
* Update banner by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2699
* fix use_reentrant by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2700
* update examples by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2703
* fix eval strategy by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2707
* Update FAQ by slin000111 in https://github.com/modelscope/ms-swift/pull/2706
* qwen to Qwen by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2708
* fix timeout & web-ui by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2709
* Fix multi lora by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2711
* support Qwen/QVQ-72B-Preview by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2712
* update examples by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2714
* fix deploy request_config by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2718
* fix examples by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2719
* fix gptq group_size by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2720
* Better error messages by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2721

New Contributors
* yrk111222 made their first contribution in https://github.com/modelscope/ms-swift/pull/2574
* lxline made their first contribution in https://github.com/modelscope/ms-swift/pull/2650

**Full Changelog**: https://github.com/modelscope/ms-swift/compare/v2.6.1...v3.0.0

New Models:
1. [Marco-o1](https://www.modelscope.cn/models/AIDC-AI/Marco-o1)
2. [mPLUG-Owl3-7B-241101](https://www.modelscope.cn/models/iic/mPLUG-Owl3-7B-241101)
3. [QwQ-32B-Preview](https://www.modelscope.cn/models/Qwen/QwQ-32B-Preview)
4. [glm-edge](ZhipuAI/glm-edge-4b-chat), [glm-edge-v](ZhipuAI/glm-edge-v-5b)

New Datasets:
1. [OpenO1-SFT](https://www.modelscope.cn/datasets/AI-ModelScope/OpenO1-SFT)

What's Changed
* support part tuner replace_key False by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2438
* bump ms version by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2449
* remove useless code by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2453
* fix qwen2-vl position_ids by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2461
* fix peft is_multimodal by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2462
* fix qwen2vl pt infer by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2463
* [TorchAcc] Update padding strategy when using persistent cache by eedalong in https://github.com/modelscope/ms-swift/pull/2464
* fix kto by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2478
* Update Common QA by slin000111 in https://github.com/modelscope/ms-swift/pull/2475
* fix awq quant device_map by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2488
* Fix preprocess num proc by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2492
* Support marco o1 by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2496
* fix eval_dataset no by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2497
* support batch flattening collator by eedalong in https://github.com/modelscope/ms-swift/pull/2499
* fix latex-ocr by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2510
* support mPLUG-Owl3 241101 by LukeForeverYoung in https://github.com/modelscope/ms-swift/pull/2515
* support qwq by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2520
* support glm-edge & glm-edge-v by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2526

New Contributors
* eedalong made their first contribution in https://github.com/modelscope/ms-swift/pull/2464

**Full Changelog**: https://github.com/modelscope/ms-swift/compare/v2.6.0...v2.6.1

English Version

Models
1. Support Qwen2.5 coder models

Feature
1. Correct and support the new loss and gradient accumulation algorithm from transformers.trainer

中文版本
模型
1. 支持千问coder系列模型

功能
1. 支持新的transformers loss和GA计算算法，并修正了其中的bug

What's Changed
* fix gptq ovis quant (inputs_embeds) by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2378
* [TorchAcc] fix qwen2 for transformers>=4.45 by baoleai in https://github.com/modelscope/ms-swift/pull/2379
* fix trl transformers==4.46 compat by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2385
* fix deploy stream media_type by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2393
* fix_mplug_owl3_cut_shape by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2394
* fix swift deploy (lmdeploy stream) by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2397
* fix lmdeploy warning & gptq-int4 support by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2401
* support qwen2.5-coder by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2400
* fix qwen_vl npu by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2408
* fix model path by yingdachen in https://github.com/modelscope/ms-swift/pull/2410
* fix qwen_vl dpo by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2411
* Compat transformers 4.46.2 loss by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2413
* Fix molmo infer by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2419
* Fix dataset map by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2421
* Support qwen2 5 coder series model by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2422
* fix PPO by hjh0119 in https://github.com/modelscope/ms-swift/pull/2377
* fix docs by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2425
* compat transformers==4.47 by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2426
* fix internvl2 lmdeploy>=0.6 deploy by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2429


**Full Changelog**: https://github.com/modelscope/ms-swift/compare/v2.5.2...v2.6.0

New Models:
1. emu3-chat
2. aya-expanse
3. ministral-8b-inst-2410

New Datasets:
1. llava-video-178k
2. moviechat-1k-test

What's Changed
* fix system by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2307
* Support Llava-video-178k dataset, a video dataset by DaozeZhang in https://github.com/modelscope/ms-swift/pull/2309
* update molmo readme by mi804 in https://github.com/modelscope/ms-swift/pull/2313
* fix molmo device_map by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2311
* support emu3-chat by mi804 in https://github.com/modelscope/ms-swift/pull/2322
* Support Moviechat-1k-test dataset, a video datatset by DaozeZhang in https://github.com/modelscope/ms-swift/pull/2319
* fix custom vllm eval args by Yunnglin in https://github.com/modelscope/ms-swift/pull/2325
* fix enable_cache by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2332
* feat: Add aya models by Aunali321 in https://github.com/modelscope/ms-swift/pull/2335
* compat transformers 4.46 by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2341
* fix streaming by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2350
* fix doc by tastelikefeet in https://github.com/modelscope/ms-swift/pull/2355
* fix internvl2 eval by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2362
* fix transformers>=4.46 loss by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2365
* Support Ministral-8b-inst-2410 by DaozeZhang in https://github.com/modelscope/ms-swift/pull/2371
* fix custom loss by Jintao-Huang in https://github.com/modelscope/ms-swift/pull/2374

New Contributors
* Aunali321 made their first contribution in https://github.com/modelscope/ms-swift/pull/2335

**Full Changelog**: https://github.com/modelscope/ms-swift/compare/v2.5.1...v2.5.2