What's Changed

• Delegate CLI script ownership to modelscope-hub by @wangxingjun778 in #1780
  The modelscope / ms console scripts are now owned solely by modelscope-hub, requires modelscope-hub>=0.3.0.
• Add Metax Dockerfile for ms-swift 4.4.x by @dwd in #1775
• Update Ascend image build and refresh runtime stack by @addsubmuldiv in #1781
• Fix studio cli by @wangxingjun778 in #1784

Full Changelog: v1.39.0...v1.40.0

---

更新内容

• CLI 脚本归属交接至 modelscope-hub by @wangxingjun778 in #1780
  modelscope / ms 命令行脚本统一由 modelscope-hub 提供，依赖 modelscope-hub>=0.3.0。
• 新增沐曦（Metax）Dockerfile，适配 ms-swift 4.4.x by @dwd in #1775
• 更新昇腾镜像构建并刷新运行时栈 by @addsubmuldiv in #1781
• 修复 studio 命令行入口 by @wangxingjun778 in #1784

完整变更：v1.39.0...v1.40.0

What's Changed

• update modelscope-hub>=0.2.0 by @wangxingjun778 in #1768

Full Changelog: v1.39.0...v1.39.1

What's Changed

• Fix some bugs by @tastelikefeet in #1746
• Fix RecursionError on Repeated Streaming Dataset Loads via Idempotent HfFileSystem Patching by @wangxingjun778 in #1754
• [Fix]dynamic module auto map double dash by @suluyana in #1761
• [Fix] Fix progress_callbacks for snapshot_download func by @wangxingjun778 in #1760
• fix: forward download kwargs in HF patcher snapshot_download calls by @suluyana in #1762

Full Changelog: v1.38.1...v1.39.0

What's Changed

• [Backport] Docker fixes from release/1.38 by @Yunnglin in #1735
• add property session in HubApi by @wangxingjun778 in #1748
• Fix release workflow hub dependency by @Yunnglin in #1749

Full Changelog: v1.38.0...v1.38.1

中文版

重构

1. 重构 Hub 与 CLI 模块，将核心 Hub 能力、下载、上传、缓存、Git 操作以及 CLI 入口迁移为基于 modelscope-hub 的兼容层，并通过插件机制保留现有命令；同时更新 modelscope-hub 依赖。更多 modelscope-hub 版本信息可参考 modelscope-hub releases。（#1732、#1733）

特性

1. 新增 ModelScope Studio 支持，提供 Studio 仓库创建、下载以及 modelscope studio CLI，可用于部署、停止、查看日志、更新设置和管理 secret。（#1727、#1742）
2. 新增私有仓库 gated_mode 支持，用户可在创建模型/数据集仓库或修改仓库可见性时启用申请制下载。（#1730）
3. 新增适用于 ms-swift 4.2.x 的 MetaX Dockerfile，并更新 Ascend NPU Dockerfile 与构建文档，完善异构硬件镜像支持。（#1734、#1736）

修复

1. 优化 trust_remote_code 与可信 owner 识别逻辑，修复官方仓库、本地目录和缓存路径下的授权判断与参数传递问题。（#1720、#1721、#1740）
2. 修复多处安全风险，包括未授权加载远程 Python 配置、执行远程脚本、不安全 eval() 解析，以及可信 owner 缓存路径解析绕过问题。（#1737、#1742）
3. 修复 create_model_tag 在网络抖动或服务端耗时时容易超时的问题，新增可配置超时、重试和退避机制。（#1723）
4. 修复 clear-cache 在单模型缓存以及自定义 MODELSCOPE_CACHE 场景下的路径解析问题。（#1724）
5. 修复 HuggingFace Hub patch 恢复逻辑及 kernels 相关测试参数缺失问题，并确保 patch_context() 在异常场景下也能正确恢复 patch 状态。（#1729、#1743）

English Version

Refactoring

1. Refactored Hub and CLI modules by moving core Hub capabilities, download, upload, cache, Git operations, and CLI entrypoints onto modelscope-hub compatibility layers, while preserving existing commands through plugins; also updated the modelscope-hub dependency. See modelscope-hub releases for more version details. (#1732, #1733)

Features

1. Added ModelScope Studio support, including Studio repository creation, downloads, and the modelscope studio CLI for deploy, stop, logs, settings updates, and secret management. (#1727, #1742)
2. Added private-repository gated_mode support, allowing users to enable application-based downloads when creating model/dataset repositories or updating repository visibility. (#1730)
3. Added MetaX Dockerfiles for ms-swift 4.2.x and updated the Ascend NPU Dockerfile and build documentation, improving heterogeneous hardware image support. (#1734, #1736)

Bug Fixes

1. Improved trust_remote_code and trusted-owner detection, fixing authorization checks and parameter propagation for official repositories, local directories, and cached model paths. (#1720, #1721, #1740)
2. Fixed multiple security risks, including unauthorized remote Python config loading, remote script execution, unsafe eval() parsing, and trusted-owner cache path spoofing. (#1737, #1742)
3. Fixed create_model_tag timeout issues under network instability or long server-side operations by adding configurable timeout, retry, and backoff handling. (#1723)
4. Fixed clear-cache path resolution for single-model cache cleanup and custom MODELSCOPE_CACHE layouts. (#1724)
5. Fixed HuggingFace Hub patch restoration and missing kernels test parameters, and ensured patch_context() restores patch state even when exceptions occur. (#1729, #1743)

Bug Fixes

1. Improved trust_remote_code and trusted-owner detection, fixing authorization checks and parameter propagation for official repositories, local directories, and cached model paths. (#1720, #1721, #1740)
2. Fixed multiple security risks, including unauthorized remote Python config loading, remote script execution, unsafe eval() parsing, and trusted-owner cache path spoofing. (#1737, #1742)
3. Fixed create_model_tag timeout issues under network instability or long server-side operations by adding configurable timeout, retry, and backoff handling. (#1723)
4. Fixed clear-cache path resolution for single-model cache cleanup and custom MODELSCOPE_CACHE layouts. (#1724)
5. Fixed HuggingFace Hub patch restoration and missing kernels test parameters, and ensured patch_context() restores patch state even when exceptions occur. (#1729, #1743)

What's Changed

• Add whitelist to iic and damo by @tastelikefeet in #1720
• Fix adaseq trust_remote_code by @tastelikefeet in #1721
• Fix timeout for create_model_tag by @wangxingjun778 in #1723
• Fix bug: 1. patch_context cannot revert classmethod 2. kernels testcase miss trust_remote_code and revision by @tastelikefeet in #1729
• 🐛 fix(clearcache): fix single model cache path by @AAAkater in #1724
• fix: quote version of timm specs in Dockerfile.ubuntu by @Seas0 in #1728
• Addgated_mode for create repo by @wangxingjun778 in #1730
• [Feature] Add studio module by @wangxingjun778 in #1727
• [Feat & Refactor] Refactor hub and CLI modules by @wangxingjun778 in #1732
• [Update] Update modelscope-hub requirement by @wangxingjun778 in #1733
• add metax dockerfile and its requirements for ms-swift 4.2.x by @WendaDeng in #1734
• Fix some security issues by @tastelikefeet in #1737
• Update npu dockerfile by @addsubmuldiv in #1736
• Fix local dir trust_remote_code by @tastelikefeet in #1740
• [Fix] Fix studio cli by @wangxingjun778 in #1742
• fix: patch_context() use try/finally to ensure unpatch on exception by @wangxingjun778 in #1743

New Contributors

• @AAAkater made their first contribution in #1724
• @Seas0 made their first contribution in #1728

Full Changelog: v1.37.1...v1.38.0

Bug Fixes:

1. Fix compatibility issues with trust_remote_code across multiple scenarios.

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.37.0...v1.37.1

中文版

特性

1. 为数据集新增基于 split 的过滤功能，支持用户按需加载指定的数据 split，并新增相应的参数校验逻辑。（#1704）
2. 增强数据集下载日志，新增调试级别日志以记录请求详情与响应指标；同时修复了数据集预览时的参数传递错误问题。（#/1700）
3. 将 Ascend NPU 的 Dockerfile 更新为基于 CANN 基础镜像构建，不再依赖 vllm-ascend 镜像，便于后续独立管理 CANN 版本与 vllm-ascend 版本。（#1687）
4. 新增适用于 ms-swift 4.1.x 的 Metax Dockerfile。（#1689）

修复

1. 修复了补丁对 kernels <= 0.12 版本的兼容性问题。（#1708）
2. 为 pipeline 和 build_trainer 函数引入 trust_remote_code 参数，要求用户显式授权后方可执行远程代码或插件，提升使用安全性。（#1703）

English Version

Features

1. Added split-based filtering for datasets, enabling users to load specific data splits on demand, along with corresponding parameter validation logic. (#1704)
2. Enhanced dataset download logging by adding debug-level logs to capture request details and response metrics; also fixed an incorrect parameter passing issue during dataset preview. (#1700)
3. Updated the Ascend NPU Dockerfile to build from the CANN base image instead of the vllm-ascend image, allowing independent version management of CANN and vllm-ascend going forward. (#1687)
4. Added a MetaX Dockerfile compatible with ms-swift 4.1.x. (#1689)

Bug Fixes

1. Fixed a compatibility issue with patches for kernels <= 0.12. (#1708)
2. Introduced the trust_remote_code parameter for the pipeline and build_trainer functions, requiring users to explicitly grant authorization before executing remote code or plugins, thereby improving usage security. (#1703)

What's Changed

• update ascend dockerfile by @addsubmuldiv in #1687
• [Fix] Fix msdatasets split issue by @wangxingjun778 in #1704
• fix patch does not support kernels<=0.12 by @tastelikefeet in #1708
• add metax dockerfile and its requirements for ms-swift 4.1.x by @WendaDeng in #1689
• [Fix] Fix dataset preview args, private streaming auth, and download retries by @wangxingjun778 in #1700
• [Fix] fix upload cache ignore by @wangxingjun778 in #1709
• Fix plugin rce by @tastelikefeet in #1703
• fix plugin by @tastelikefeet in #1712
• fix trainer pass trust_remote_code to model by @tastelikefeet in #1713
• [Fix] fix invalid commit by @wangxingjun778 in #1715
• [Fix] Fix split detection by @wangxingjun778 in #1714
• update npu dockerfile by @xvxuopop in #1716
• Fix pipeline parameter by @tastelikefeet in #1717

New Contributors

• @WendaDeng made their first contribution in #1689
• @xvxuopop made their first contribution in #1716

Full Changelog: v1.36.3...v1.37.0

Release Notes

中文版

修复 (Fix)

• CLI 上传报告：修复上传计数逻辑，优化报告生成流程 (#1694)
• CLI 帮助文档：统一 --endpoint 参数描述，移除冗余站点标签 (#1696)
• 创建模型仓库：为 create 命令添加 endpoint 支持，新增 --exist_ok 标志；移除冗余的文件大小检查逻辑 (#1699)

特性 (Feature)

• CLI endpoint 重构：统一 login/create/upload/download 命令的 endpoint 解析逻辑，新增 resolve_endpoint 工具函数，支持 URL 自动补全与环境变量回退 (#1695)
• Kernels 下载支持：通过 monkey-patching 机制，支持从 ModelScope 下载 kernels 库，兼容 HuggingFace 接口 (#1697)

---

English Version

Fixes

• Upload Report: Fixed upload count logic and optimized report generation flow (#1694)
• CLI Help Docs: Unified --endpoint parameter description across commands; removed redundant site labels (#1696)
• Model Repo Creation: Added endpoint support and --exist_ok flag to create command; removed redundant non-LFS file size checks (#1699)

Features

• CLI Endpoint Refactor: Standardized endpoint resolution for login/create/upload/download commands with new resolve_endpoint utility; supports URL auto-completion and env var fallbacks (#1695)
• Kernels Download Support: Enabled downloading kernels via ModelScope using monkey-patching for HuggingFace API compatibility (#1697)

---

What's Changed

• [Fix] Upload report by @wangxingjun778 in #1694
• [Feat & Fix] Refactor endpoint arg for CLI by @wangxingjun778 in #1695
• [Fix] Update helper desc by @wangxingjun778 in #1696
• Support kernels downloading by @tastelikefeet in #1697
• [Fix] Add endpoint for creating model repo by @wangxingjun778 in #1699

Full Changelog: v1.36.2...v1.36.3

Release Notes (English)

Authentication & API

• [Fix] OpenAPI Auth Support (#1686)
  • Centralized bearer token handling via _build_bearer_headers in HubApi
  • Unified response parsing with _parse_openapi_response for maintainability
  • Enhanced robustness: null-safe URL parsing, JSON decoding error handling, and response type validation

Upload Module

• [Refactor&Fix] Upload Folder Module (#1690)
  • Introduced UploadTracker for resumable uploads and progress tracking
  • Added adaptive batching and "ReAct" progressive retry strategy
  • Application-level retries for blob uploads with improved error classification
  • File size integrity checks and UTF-8 encoding for cross-platform compatibility
  • Fixed retry backoff calculation and upload hang issues

---

版本更新说明（中文）

认证与API

• [修复] OpenAPI 认证支持 (#1686)
  • 在 HubApi 基类中统一实现 bearer token 认证逻辑
  • 统一响应解析方法，提升代码可维护性
  • 增强健壮性：空值安全的URL解析、JSON解码异常兼容处理、响应类型校验

上传模块

• [重构&修复] 上传文件夹模块优化 (#1690)
  • 新增 UploadTracker 管理器，支持断点续传与进度追踪
  • 支持自适应批处理与"ReAct"渐进式重试兜底策略
  • 为blob上传增加应用层重试机制，优化错误分类
  • 新增文件大小校验，强制UTF-8编码确保跨平台兼容
  • 修复重试退避时间计算逻辑及上传卡死问题

---

What's Changed

• [Feature] Add openapi auth support by @wangxingjun778 in #1686
• [Refactor&Fix] Upload folder module refactor by @wangxingjun778 in #1690

Full Changelog: v1.36.1...v1.36.2

📦 Release Notes (English)

ModelScope Download Module Refactor

• Architecture Upgrade: Rebuilt download module with producer-consumer pipeline for significantly improved throughput and responsiveness.
• Memory Optimization: Added server-side prefix filtering + paginated listing to reduce memory footprint for large-scale snapshots.
• Reliability Enhancements:
  • Retry mechanism for hash validation failures
  • Larger buffer size for hash computation
  • Fault-tolerant reporting with strict exception handling for pagination/partial-download errors
• Error Handling: Introduced FileDownloadError for granular single-file download failure tracking.
• Config Updates:
  • Default dataset page size set to 200
  • Updated build dependencies (pip, setuptools)

---

📦 发布说明（中文）

ModelScope 下载模块重构

• 架构升级：采用生产者 - 消费者流水线重构下载模块，显著提升下载吞吐与响应速度。
• 内存优化：支持服务端 prefix 过滤 + 分页列表拉取，大幅降低大规模快照下载时的内存占用。
• 可靠性增强：
  • 哈希校验失败自动重试
  • 增大哈希计算缓冲区
  • 容错上报机制，确保分页错误/部分下载失败时严格抛出异常，避免快照状态不一致
• 错误处理：新增 FileDownloadError 异常类，支持单文件下载失败的精细化追踪。
• 配置更新：
  • 默认数据集分页大小调整为 200
  • 更新构建依赖（pip, setuptools）

---

💡 Migration Tip: This is a backward-compatible refactor. No user code changes required, but users will notice faster and more stable downloads, especially for large-scale datasets.

What's Changed

• [Refactor] Refactor the modelscope download module by @wangxingjun778 in #1683

Full Changelog: v1.36.0...v1.36.1

中文版

✨ 新特性

• 文件夹上传优化：新增生产者-消费者管道机制，并发上传与进度追踪，大幅提升超大型数据集上传效率（#1671）

问题修复

• CSV 预览加载：修复流式加载 CSV 时的兼容性问题，支持 engine=python 并修复字段长度溢出错误（#1665）
• Swift 4.0 兼容：适配 ms-swift 4.0 包结构调整，更新模块导入路径，保持向后兼容（#1680）

部署

• GPG 密钥修复：更新 Dockerfile 中 GPG KEY 配置，支持多密钥验证（#1678）
• Ascend 镜像优化：切换至预构建 vllm-ascend 基础镜像，精简构建步骤，加快镜像生成速度（#1679）

---

🇬🇧 English Version

✨ Features

• Folder Upload Enhancement: Introduced producer-consumer pipeline for folder uploads with concurrent management, and progress tracking, significantly improving large-scale upload efficiency (#1671)

Bug Fixes

• CSV Preview Loading: Fixed streaming CSV loading issues, added engine=python support and resolved field size OverflowError (#1665)
• Swift 4.0 Compatibility: Updated import paths to align with ms-swift 4.0 package restructuring while maintaining backward compatibility (#1680)

Deployment

• GPG KEY Fix: Updated Dockerfile GPG KEY configuration to support multi-key verification (#1678)
• Ascend Dockerfile Optimization: Migrated to prebuilt vllm-ascend base image and streamlined build process for faster image generation (#1679)

---

What's Changed

• [Feature] Add producer-consumer pipeline for uploading folder by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1671
• [Fix] Fix preview csv stream loading by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1665
• Fix GPG KEY by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1678
• [bugfix] compat swift 4.0 by @Jintao-Huang in https://github.com/modelscope/modelscope/pull/1680
• update ascend dockerfile by @addsubmuldiv in https://github.com/modelscope/modelscope/pull/1679

New Contributors

• @addsubmuldiv made their first contribution in https://github.com/modelscope/modelscope/pull/1679

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.35.4...v1.36.0

📝 Release Notes / 发布说明

🇨🇳 Chinese (中文)

新功能 (Features)

• 支持 Collections 功能：引入了对模型/数据集集合（Collections）的支持，方便用户更好地组织和管理资源。 (#1656)

修复与优化 (Fixes & Improvements)

• 提升下载 CLI 健壮性：增强了 ModelScope 命令行下载工具的稳定性，改善用户体验。 (#1662)
• 修复下载 Token 认证问题：解决了在下载过程中可能出现的 Token 认证失败问题，确保权限验证正常。 (#1674)
• 修复 Skill 目录拼接错误：修正了 Skill 模块中目录路径拼接的逻辑错误。 (#1657)
• 优化预览加载与 HF 数据集工具：修复了预览加载问题，并对 Hugging Face 数据集相关工具类进行了重构，提升代码质量和可维护性。 (#1654)

---

🇺🇸 English

Features

• Collections Support: Added support for Collections, allowing users to better organize and manage models and datasets. (#1656)

Fixes & Improvements

• Enhanced Download CLI Robustness: Improved the stability and robustness of the ModelScope command-line download tool. (#1662)
• Fixed Token Authentication for Downloads: Resolved issues related to token authentication during the download process to ensure secure and successful access. (#1674)
• Fixed Skill Directory Concatenation: Corrected a bug in the directory path concatenation logic within the Skill module. (#1657)
• Preview Loading Fix & HF Utils Refactor: Fixed issues with preview loading and refactored Hugging Face dataset utilities for better code structure and maintainability. (#1654)

What's Changed

• merge 1.35.1 to master by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1653
• Feat/collections by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1656
• [Fix] Skill DIR concat by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1657
• [Fix | Refactor] Fix preview loading and refactor hf datasets utils by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1654
• [Fix] Enhance the robustness of ModelScope download cli by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1662
• [Fix] fix token auth for downloading by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1674

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.35.3...v1.35.4

中文版本

新特性

1. 提升与 datasets 库的兼容性，适配了 4.0 及以上版本的变更，修复了因文件大小获取错误导致的数据预览加载问题，同时重构了代码结构并优化了 API 缓存机制以提升性能。

English Version

New Feature

1. Improves compatibility with datasets library 4.0+, fixes dataset preview loading issues caused by incorrect file size reporting, and refactors the codebase with optimized API caching for better performance.

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.35.2...v1.35.3

中文版本

新特性

1. 支持skills的下载

# Skills will be default placed to `~/.agents/skills`
# Download a collection of skills
modelscope download --collection MiniMax/MiniMax-Office-skills [--token xxx] [--local_dir ./temp]
# Download a batch skills of a single skill
modelscope skills add @MiniMax-AI/minimax-pdf @xxx/xxx [--token xxx]

English Version

New Feature

1. Support the downloading of skills

# Skills will be default placed to `~/.agents/skills`
# Download a collection of skills
modelscope download --collection MiniMax/MiniMax-Office-skills [--token xxx] [--local_dir ./temp]
# Download a batch skills of a single skill
modelscope skills add @MiniMax-AI/minimax-pdf @xxx/xxx [--token xxx]

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.35.1...v1.35.2

中文版本

Bug修复

1. 修复了oss2被全局依赖的问题
2. 修复了packaging没有被加入依赖的问题

English Version

BugFix

1. Fix an issue that oss2 was required by global import.
2. Fix an issue that packaging was not included in the dependencies.

What's Changed

• [Fix] lazy import oss2 by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1649
• fix packaging dependency by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1651

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.35.0...v1.35.1

中文版本

新特性

1. ms_dataset模块重构，适配最新的datasets依赖（4.x大版本）
2. HubApi中的upload_file和upload_folder功能，增加对数据分块的重试功能
3. 重写镜像打包逻辑，将原有cpu、gpu镜像替换为old版本进行维护；原有的LLM镜像扩展出一个CPU镜像作为主镜像；同时原有swift镜像替换为latest镜像，表示依赖为最新版的镜像
4. 更换了CI镜像为Stable GPU镜像（原LLM镜像）

Bug修复

1. 修复了跨模型加载Config时，下载了全模型的问题
2. 修复了下载config和tokenizer文件时，正则表达式无效把模型文件也下载下来的问题
3. 修复pkg_resources问题
4. 修复了transformers5.0带来的CI报错，跳过了不兼容的小模型CI用例

English Version

New Features

1. Refactored the ms_dataset module to support the latest datasets dependency (major version 4.x)
2. Added retry functionality for chunked uploads in the upload_file and upload_folder features of HubApi
3. Rewrote the image packaging logic: replaced the original CPU and GPU images with "old" versions for maintenance; created a new CPU image from the original LLM image to serve as the main image; renamed the original swift image to "latest" to indicate it contains the most up-to-date dependencies
4. Switched the CI image to the Stable GPU image (formerly the LLM image)

Bug Fixes

1. Fixed an issue where the entire model was downloaded when loading Config across different models
2. Fixed an issue where an invalid regular expression caused model files to be downloaded along with config and tokenizer files
3. Fixed the pkg_resources issue
4. Fixed CI errors caused by transformers 5.0 by skipping incompatible small model CI test cases

---

What's Changed

• Fix #1316 by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1612
• [Fix] Fix aigc cover image by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1613
• fix downloading txt files by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1610
• lint code by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1619
• Fix downloading repos in automap by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1630
• fix unitlock by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1635
• compat with tf5.0 by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1618
• [Bugfix] Make local_files_only configurable by @Potabk in https://github.com/modelscope/modelscope/pull/1632
• Feat/update docker file by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1640
• [Feat] dataset module refactor by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1623
• add metax dockerfile and its requirements for ms-swift by @vx120 in https://github.com/modelscope/modelscope/pull/1643
• [Fix] fix dataset util by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1645

New Contributors

• @Potabk made their first contribution in https://github.com/modelscope/modelscope/pull/1632
• @vx120 made their first contribution in https://github.com/modelscope/modelscope/pull/1643

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.34.0...v1.35.0

中文版

• HubApi 认证优化 (#1573): 更新了 HubAPI 的 cookie 获取逻辑，支持在初始化时传入 token 参数，并优化了 token 的获取优先级（函数参数 > 实例属性 > 环境变量）。
• 修复并完善了模型与数据集操作中 token 参数的透传逻辑 (#1592, #1593): 统一认证令牌传递机制，确保 API 调用携带有效认证凭据。同时优化了 Cookie 处理逻辑，提升与 ModelScope Hub 交互的可靠性与安全性。
• 修复 OSS STS 认证过期问题 (#1589): 重构了 OSS 工具类，新增 CredentialProviderWrapper 来自动处理临时凭证的刷新，解决了长时间数据集下载时 STS token 过期导致的失败问题。
• 弃用仓库删除相关 API (#1588): 出于安全考虑，暂时弃用 delete_repo、delete_model 和 delete_dataset 方法。这些方法现在会发出 DeprecationWarning 警告，将在未来版本中基于完善的 token 认证机制重新启用。
• 修复 FSMN 模型 ONNX 导出问题 (#1576, #1577): 为 UniDeepFsmnDilated 模块的 squeeze 操作显式指定 dim 参数，修复了 ONNX 导出后在 TensorRT 中运行失败的问题。
• 修复 Docker 镜像构建的 APT 源问题 (#1575): 移除 Dockerfile 中的 nvidia.cn APT 源，解决了在某些网络环境下 apt-get update 失败的问题。
• 修复数据集下载的依赖问题 (#1581): 将 is_relative_path 函数从 datasets 库的导入改为使用内置实现，解决了与特定版本 datasets 库的兼容性问题。
• 修复 ignore_patterns 重置问题 (#1578): 修复了当 allow_patterns 包含 ** 时错误地清空 ignore_patterns 的问题。

English Version

• HubAPI authentication enhancement (#1573): Updated HubAPI cookie retrieval logic to support passing a token parameter during initialization, with optimized token priority (function parameter > instance attribute > environment variable).
• Fixed and improved token propagation in model and dataset operations (#1592, #1593): Unified authentication token handling to ensure API calls carry valid credentials. Also enhanced cookie handling logic to improve reliability and security when interacting with ModelScope Hub.
• Fixed OSS STS authentication expiry issue (#1589): Refactored OSS utility classes by introducing CredentialProviderWrapper to automatically handle temporary credentials refresh, resolving failures during long-running dataset downloads due to STS token expiration.
• Deprecated repository deletion APIs (#1588): For security reasons, temporarily deprecated delete_repo, delete_model, and delete_dataset methods. These methods now emit DeprecationWarning and will be re-enabled in future versions with proper token authentication mechanisms.
• Fixed FSMN model ONNX export issue (#1576, #1577): Explicitly specified the dim parameter for the squeeze operation in UniDeepFsmnDilated module, fixing failures when running ONNX-exported models in TensorRT.
• Fixed Docker image build APT source issue (#1575): Removed nvidia.cn APT sources from Dockerfile to fix apt-get update failures in certain network environments.
• Fixed dataset download dependency issue (#1581): Replaced is_relative_path import from the datasets library with a built-in implementation, resolving compatibility issues with specific versions of the datasets library.
• Fixed ignore_patterns reset issue (#1578): Fixed an issue where ignore_patterns was incorrectly cleared when allow_patterns contained **.

What's Changed

• Feat: update hubapi get cookie by @Yunnglin in https://github.com/modelscope/modelscope/pull/1573
• Fix: remove reset ignore_patterns by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1578
• Fix: datasets dependency for snapshot download by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1581
• fix: Remove nvidia.cn from APT sources to fix apt failures in certain… by @suluyana in https://github.com/modelscope/modelscope/pull/1575
• Fix TensorRT parse failure for MossFormer2 ONNX Squeeze axes by @Jzz1943 in https://github.com/modelscope/modelscope/pull/1577
• Fix: deprecate delete_repo, delete_model and delete_dataset due to token a… by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1588
• [Fix]oss utils sts auth expire issue by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1589
• [Fix] hub push model with token by @Yunnglin in https://github.com/modelscope/modelscope/pull/1592
• [Fix] HubAPI token usage by @Yunnglin in https://github.com/modelscope/modelscope/pull/1593

New Contributors

• @Jzz1943 made their first contribution in https://github.com/modelscope/modelscope/pull/1577

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.33.0...v1.34.0

中文版

主要更新

• 更新默认分支版本（default revision）策略，以提升仓库版本管理的一致性与易用性 #1546 #1553
• Hub API 新增 set_repo_visibility 接口，支持设置模型仓库可见性，并补充可见性参数校验和错误日志提示 #1564 #1568 #1569 #1570
• 新增 Ascend NPU 的 Dockerfile，便于在昇腾 NPU 环境中部署与使用 ModelScope #1550
• snapshot_download 新增 DEFAULT_MAX_WORKERS 参数配置，优化模型快照下载的并发行为 #1566

AIGC 相关能力增强

• 为 AIGC 的 create_model_tag 接口新增参数，支持更灵活的模型标签创建与管理 #1555
• AIGC 模型新增 base_model_sub_type 字段，便于更精细地描述与区分底座模型类型 #1563

Hub 与上传下载体验优化

• 修正 upload_folder 的默认忽略规则，避免重要文件被错误忽略 #1549
• 修复上传模式匹配（upload patterns）相关问题，提升文件上传的准确性 #1567

稳定性与工程改进

• 修复影响 hasattr 行为的实验类型定义问题，避免潜在属性检测异常 #1552
• 动态模块加载逻辑补丁修复，提升动态模块相关功能的稳定性 #1554
• 更新构建文件与 CI 工作流，改进整体构建发布流程 #1551

English Version

Key Updates

• Updated default revision strategy to improve consistency and usability of repository version management #1546 #1553
• Added set_repo_visibility in Hub API to control repository visibility, with input validation and detailed error logging #1564 #1568 #1569 #1570
• Added Ascend NPU Dockerfile to simplify deployment on Ascend NPU environments #1550
• Introduced DEFAULT_MAX_WORKERS in snapshot_download to better control concurrent download behavior #1566

AIGC Enhancements

• Added new arguments for AIGC create_model_tag for more flexible model tag creation and management #1555
• Added base_model_sub_type for AIGC models to describe and distinguish base model types more precisely #1563

Hub & Upload/Download Improvements

• Fixed default ignore patterns in upload_folder to avoid unintentionally skipping important files #1549
• Fixed upload patterns issues to improve accuracy of file uploads #1567

Stability & Engineering

• Fixed experiment type definition that affected hasattr behavior and could cause attribute detection issues #1552
• Patched dynamic module handling to improve stability of dynamic module related features #1554
• Updated build files and CI workflows to refine the overall build and release pipeline #1551

What's Changed

• Feat/update default revision by @alcholiclg in https://github.com/modelscope/modelscope/pull/1546
• Fix default ignore patterns for upload_folder by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1549
• Fix exp type which affects hasattr by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1552
• Patch dynamic module by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1554
• Feat: add new args for aigc create_model_tag by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1555
• [docker] Add Ascend NPU dockerfile by @baymax591 in https://github.com/modelscope/modelscope/pull/1550
• Update build file and workflow by @Yunnglin in https://github.com/modelscope/modelscope/pull/1551
• [WIP]Feat/update default revision by @alcholiclg in https://github.com/modelscope/modelscope/pull/1553
• Feat: add set_repo_visibility in hub api by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1564
• Add base_model_sub_type for aigc model by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1563
• Feat: add DEFAULT_MAX_WORKERS in snapshot_download func by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1566
• Fix/upload patterns by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1567
• fix set_repo_visibility by @suluyana in https://github.com/modelscope/modelscope/pull/1568
• Fix: add error log for invalid input visibility args by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1569
• fix: check tasks in HubApi.set_repo_visibility by @suluyana in https://github.com/modelscope/modelscope/pull/1570

New Contributors

• @baymax591 made their first contribution in https://github.com/modelscope/modelscope/pull/1550

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.32.0...v1.33.0

中文版

1. 支持使用 pyproject.toml 来安装 modelscope 库，保证与之前安装方式兼容 #1542
2. hub api 的 list_datasets 接口切换为 OpenAPI 端点 #1532
3. AIGC 模型支持 model_source 参数 #1530
4. ModelScope Hub 添加定期自动上传功能：自动定期上传本地文件夹到 ModelScope Hub，增量提交日志、训练产物等文件 #1480
5. AIGC 的 cover_images 支持 mp4 文件 #1533
6. 为 .safetensors 等单个模型文件添加内部云加速 #1529

English Version

1. Support using pyproject.toml to install the modelscope library, ensuring compatibility with previous installation methods #1542
2. Switch the list_datasets interface of hub API to OpenAPI endpoint #1532
3. AIGC models support the model_source parameter #1530
4. Add scheduled automatic upload functionality to ModelScope Hub: automatically upload local folders to ModelScope Hub at regular intervals, with incremental commits of logs, training artifacts, and other files #1480
5. AIGC's cover_images now support mp4 files #1533
6. Add internal cloud acceleration for single model files such as .safetensors #1529

What's Changed

• Feat/support commit scheduler by @alcholiclg in https://github.com/modelscope/modelscope/pull/1480
• Add INTRA_CLOUD_ACCELERATION for single model file downloading by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1529
• Feat: add model source for aigc model by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1530
• Add trust_remote_code for _download_additional_modules by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1521
• Update list datasets to OpenAPI by @Yunnglin in https://github.com/modelscope/modelscope/pull/1532
• [Feature] Add pyproject.toml by @Yunnglin in https://github.com/modelscope/modelscope/pull/1542
• Remove raise FileIntegrityError by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1517
• [WIP]change DEFAULT_MODEL_REVISION to master by @alcholiclg in https://github.com/modelscope/modelscope/pull/1544
• update ollama: granite4,deepseek-v3.1,gpt-oss by @xiaojianpinga in https://github.com/modelscope/modelscope/pull/1531
• Add mp4 support for AIGC cover_images by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1533
• fix daily regression by @Yunnglin in https://github.com/modelscope/modelscope/pull/1545

New Contributors

• @xiaojianpinga made their first contribution in https://github.com/modelscope/modelscope/pull/1531

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.31.0...v1.32.0

English Version

1. Integrate support for "official tags" of AIGC models. https://github.com/modelscope/modelscope/pull/1509
2. Extend the network timeout duration for model_revision checking. https://github.com/modelscope/modelscope/pull/1507
3. Update the ms-swift Docker build process. https://github.com/modelscope/modelscope/pull/1502
4. Fix some bugs.

中文版

1. 集成对 AIGC 模型“官方标签”的支持。 https://github.com/modelscope/modelscope/pull/1509
2. 延长 model_revision 检查的网络超时时间。 https://github.com/modelscope/modelscope/pull/1507
3. 更新 ms-swift Docker 构建流程。 https://github.com/modelscope/modelscope/pull/1502
4. 修复了一些bugs。

What's Changed

• [docker] update swift docker by @Jintao-Huang in https://github.com/modelscope/modelscope/pull/1502
• Update timeout for checking model revision by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1507
• Fix get_domain and endpoint by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1511
• Fix SyntaxError warning on MODELSCOPE_ASCII by @MichaelCurrie in https://github.com/modelscope/modelscope/pull/1512
• [fix] ans_dfsmn_pipeline.py resample by @weedge in https://github.com/modelscope/modelscope/pull/1510
• fix tokenizer download by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1516
• Feat/add official tag by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1509

New Contributors

• @MichaelCurrie made their first contribution in https://github.com/modelscope/modelscope/pull/1512
• @weedge made their first contribution in https://github.com/modelscope/modelscope/pull/1510

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.30.0...v1.31.0

English Version

1. Added support for creating model tags via API or CLI. (@Koko-ry, #1487)
2. Extended repo_info functionality to include the siblings field, enabling better metadata inspection. (@alcholiclg, #1490)
3. Support for uploading AIGC model cover images in Base64 format. (@Koko-ry, #1492)
4. Fixed compatibility issues with models when using the latest version of Hugging Face Transformers. (@tastelikefeet, #1491)
5. Resolved linting errors in utility modules to improve code quality and consistency. (@wangxingjun778, #1493)
6. Corrected endpoint behavior for AI-generated content (AIGC) services. (@Koko-ry, #1494)
7. Fixed invalid revision reference in repo_info responses, ensuring accurate version tracking. (@alcholiclg, #1496)
8. Improved model tag descriptions by correcting formatting and enhancing clarity. (@Koko-ry, #1500)

中文版本

1. 新增通过 API 或 CLI 创建模型标签的功能。(@Koko-ry, #1487)
2. 扩展 repo_info 功能，支持返回 siblings 字段，提升元数据查看能力。(@alcholiclg, #1490)
3. 支持使用Base64格式上传AIGC模型的封面图。 (@Koko-ry, #1492)
4. 修复与最新版 Hugging Face Transformers 的模型兼容性问题。(@tastelikefeet, #1491)
5. 修复工具模块中的代码规范（lint）错误，提升代码质量与一致性。(@wangxingjun778, #1493)
6. 修正 AI 生成内容（AIGC）服务端点的行为逻辑。(@Koko-ry, #1494)
7. 修复 repo_info 响应中无效的版本号引用，确保版本追踪准确无误。(@alcholiclg, #1496)
8. 优化模型标签描述内容，修正格式并提升可读性。(@Koko-ry, #1500)

What's Changed

• fix models with the latest transformers by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1491
• add create model tag by @Koko-ry in https://github.com/modelscope/modelscope/pull/1487
• add support for siblings field in repo_info by @alcholiclg in https://github.com/modelscope/modelscope/pull/1490
• Fix/upload base64 by @Koko-ry in https://github.com/modelscope/modelscope/pull/1492
• Fix utils lint by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1493
• fix aigc endpoint by @Koko-ry in https://github.com/modelscope/modelscope/pull/1494
• Fix invalid revision in repo_info by @alcholiclg in https://github.com/modelscope/modelscope/pull/1496
• Fix model tag des by @Koko-ry in https://github.com/modelscope/modelscope/pull/1500

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.29.2...v1.30.0

English Version

1. Support repo_info support for retrieving repository-level metadata (datasets, models) and list all commit info
2. Fix some bugs

中文版本

1. 支持获取repo级别的详细信息，以及获取所有的提交记录
2. 修复了一些bug

What's Changed

• Fix/upload commit by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1459
• Fix: add error log for validate_blob and upload_blob by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1482
• Fix:save_cached_files for sync dumping by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1484
• Feat/support repo info by @alcholiclg in https://github.com/modelscope/modelscope/pull/1483

New Contributors

• @alcholiclg made their first contribution in https://github.com/modelscope/modelscope/pull/1483

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.29.1...v1.29.2

English Version

1. Support arg parser when python version >= 3.12, thanks to the contribution of @yaqiangsun
2. Fix some bugs, e.g. VLLM cannot download from ModelScope of deepseek/internvl/internlm models

中文版本

1. 兼容python>=3.12时的arg parser，感谢@yaqiangsun的贡献
2. 修复了一些问题，例如VLLM无法从ModelScope下载deepseek和浦江实验室的模型

What's Changed

• Fix intermixed missing in cli_argument_parser.py by @yaqiangsun in https://github.com/modelscope/modelscope/pull/1437
• Fix/aigc weight by @Koko-ry in https://github.com/modelscope/modelscope/pull/1464
• Fix trust_remote_code by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1462
• fix:file already exist by @Koko-ry in https://github.com/modelscope/modelscope/pull/1465
• Merge 1.29 to master by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1469
• support patch intern and deepseek by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1475
• skip aigc test by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1476

New Contributors

• @yaqiangsun made their first contribution in https://github.com/modelscope/modelscope/pull/1437

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.29.0...v1.29.1

English Version

New Features

1. Added support for AIGC model creation interface:

modelscope create <repo_id> --token <token> --visibility <public/private/internal> --aigc --model_path <model_path> --aigc_type <Checkpoint/LoRA/VAE> --base_model_type <base_model_type>

1. Added support for using ModelScope MCP server. Users can interface with MCP plaza through the following method:

from modelscope.hub.mcp_api import MCPApi
server = MCPApi().get_mcp_server('@modelcontextprotocol/fetch')
...

1. Added support for repository creation command:

modelscope create owner/repo_name --token ms-xxx --repo_type model --visibility private --chinese_name 测试模型1 --license MIT

Bug Fixes

1. Fixed some security issues
2. Fixed issues where some models could not run on higher Python versions. These models will now throw exceptions directly, guiding users to use lower Python versions
3. Added additional trust_remote_code validation for models that need to run external code or use torch.load during loading

中文版本

新功能

1. 支持了AIGC模型的创建接口:

modelscope create <repo_id> --token <token> --visibility <public/private/internal> --aigc --model_path <model_path> --aigc_type <Checkpoint/LoRA/VAE> --base_model_type <base_model_type>

1. 增加了对ModelScope MCP server的使用支持，用户可以通过多个接口与MCP广场交互，例如：

from modelscope.hub.mcp_api import MCPApi
server = MCPApi().get_mcp_server('@modelcontextprotocol/fetch')
...
...

1. 支持了创建repo的命令：

modelscope create owner/repo_name --token ms-xxx --repo_type model --visibility private --chinese_name 测试模型1 --license MIT

Bug修复

1. 修复了一些安全问题
2. 修复了部分模型在高版本python下无法运行的问题，这些模型会直接抛出异常，指引用户使用更低版本的python
3. 对部分模型在加载时需要运行外部code或者使用torch.load的情况增加了额外的trust_remote_code判断

What's Changed

• fix pyyaml according to: https://github.com/Anchor0221/CVE-2025-50460 by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1428
• 解决使用uv tool的依赖的问题 by @Tian-Jionglu in https://github.com/modelscope/modelscope/pull/1427
• Merge release 1.28 by @hjh0119 in https://github.com/modelscope/modelscope/pull/1401
• ollama template: qwen3-coder, gemma3n, mistral-small3.2, magistral, devstral by @suluyana in https://github.com/modelscope/modelscope/pull/1434
• add ms-agent into dockerfile by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1436
• fix large log in scan cache test by @Yunnglin in https://github.com/modelscope/modelscope/pull/1439
• Fix: add revision arg for upload_file by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1440
• fix bug by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1441
• Feat: add atomic capabilities - MCPApi by @Koko-ry in https://github.com/modelscope/modelscope/pull/1426
• Add create repo cli by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1445
• feat: sentence_embedding pipeline by @suluyana in https://github.com/modelscope/modelscope/pull/1435
• Feat: Add AIGC model creation support to create_model by @Koko-ry in https://github.com/modelscope/modelscope/pull/1431
• merge release1.28 by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1448
• fix: TestTinynasDamoyoloTrainerSingleGPU ut by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1449
• fix: fix mcp api ut by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1450
• fix: fix ci by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1452
• fix tensorflow cannot load gpu by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1456
• Update default version of python by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1457
• Add create cli warning by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1458

New Contributors

• @Tian-Jionglu made their first contribution in https://github.com/modelscope/modelscope/pull/1427

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.28.2...v1.29.0

V1.28.2 Release Note

English Version

1. Hotfix datasets features for datasets >=3.0.0, <=3.6.0 #1447

中文版本

1. 修复datasets features兼容性问题 #1447

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.28.1...v1.28.2

V1.28.1 Release Note

English Version

1. Fix use_cookies for get_model_files in delete_files: https://github.com/modelscope/modelscope/pull/1424
2. Add MODELSCOPE_CREDENTIALS_PATH env for user-defined authorization path: https://github.com/modelscope/modelscope/pull/1423
3. Skip UT TextToSpeechSambertHifigan16kPipelineTest: https://github.com/modelscope/modelscope/pull/1420
4. Add weak file lock for snapshot_download: https://github.com/modelscope/modelscope/pull/1417
5. Fix RCE issue for plugins (add trust_remote_code) for from_pretrained function: https://github.com/modelscope/modelscope/pull/1415
6. Fix get_dataset_infos to avoid list all pages for dataset: https://github.com/modelscope/modelscope/pull/1414

中文版本

1. 修复delete_files函数中的get_model_files调用，当需要鉴权的场景，需要明确传入use_cookies参数 : https://github.com/modelscope/modelscope/pull/1424
2. 增加环境变量MODELSCOPE_CREDENTIALS_PATH，支持用户自定义配置本地授权的存储路径（cookies、session等信息）: https://github.com/modelscope/modelscope/pull/1423
3. 跳过测试用例TextToSpeechSambertHifigan16kPipelineTest（用例失效）: https://github.com/modelscope/modelscope/pull/1420
4. 在snapshot_download中增加weak file lock: https://github.com/modelscope/modelscope/pull/1417
5. 在from_pretrained中，增加 trust_remote_code`参数，避免插件RCE问题: https://github.com/modelscope/modelscope/pull/1415
6. 修复get_dataset_infos函数中的分页问题：https://github.com/modelscope/modelscope/pull/1414

What's Changed

• Fix use_cookies in get_model_files by @wangxingjun778 #1424
• Add MODELSCOPE_CREDENTIALS_PATH for user-defined authorization path by @wangxingjun778 #1423
• Skip UT TextToSpeechSambertHifigan16kPipelineTest temporarily by @Koko-ry #1420
• weak file lock by @tastelikefeet #1417
• Fix RCE issue for plugins by @Koko-ry #1415
• Fix dataset infos by @wangxingjun778 #1414

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.28.0...v1.28.1

English Version

• Added Hub download progress callback: Supports custom callback functions for download progress updates. https://github.com/modelscope/modelscope/pull/1380
• Added environment variable configuration: Introduced the SIZE_THRESHOLD_TO_ENFORCE_LFS environment variable to configure the file size threshold for enforcing LFS usage. https://github.com/modelscope/modelscope/pull/1383
• Added wildcard batch deletion: Supports batch deletion of files in model or dataset repositories using wildcards. https://github.com/modelscope/modelscope/pull/1384
• Optimized VAE processing: Improved VAE chunking workflow and reduced VRAM consumption. https://github.com/modelscope/modelscope/pull/1378
• Added Hub upload environment variables: Added environment variable configurations for Hub uploads. https://github.com/modelscope/modelscope/pull/1389
• Compatibility improvement: Added compatibility support for datasets library version 3.6. https://github.com/modelscope/modelscope/pull/1393
• Bug fix: Resolved an issue where AutoProcessor and AutoTokenizer might download incomplete files when using from_pretrained. https://github.com/modelscope/modelscope/pull/1398

中文版本

• 新增 Hub 下载进度回调支持： 支持自定义下载进度更新回调函数。 https://github.com/modelscope/modelscope/pull/1380
• 新增环境变量配置： 添加 SIZE_THRESHOLD_TO_ENFORCE_LFS 环境变量，用于配置强制使用 LFS 的文件大小阈值。 https://github.com/modelscope/modelscope/pull/1383
• 新增通配符批量删除功能： 支持使用通配符批量删除模型或数据集仓库中的文件。 https://github.com/modelscope/modelscope/pull/1384
• 优化 VAE 处理： 优化 VAE 分块处理流程并减少显存占用。https://github.com/modelscope/modelscope/pull/1378
• 新增 Hub 上传环境变量配置： 增加 Hub 上传相关的环境变量配置选项。 https://github.com/modelscope/modelscope/pull/1389
• 兼容性提升： 兼容 datasets 库 3.6 版本。 https://github.com/modelscope/modelscope/pull/1393
• 问题修复： 修复 AutoProcessor 和 AutoTokenizer 使用 from_pretrained 方法时可能出现的文件下载不完整问题。https://github.com/modelscope/modelscope/pull/1398

What's Changed

• [hub] support ProgressCallback by @Jintao-Huang in https://github.com/modelscope/modelscope/pull/1380
• Add env SIZE_THRESHOLD_TO_ENFORCE_LFS by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1383
• Feat deletefiles by @Koko-ry in https://github.com/modelscope/modelscope/pull/1384
• Merge release 1.27 by @hjh0119 in https://github.com/modelscope/modelscope/pull/1392
• perf(vae): optimize tiling performance and memory usage by @MengAiDev in https://github.com/modelscope/modelscope/pull/1378
• Add upload check envs by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1389
• Upgrade datasets by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1393
• Fix ignore files of AutoProcessor by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1398

New Contributors

• @Koko-ry made their first contribution in https://github.com/modelscope/modelscope/pull/1384
• @hjh0119 made their first contribution in https://github.com/modelscope/modelscope/pull/1392
• @MengAiDev made their first contribution in https://github.com/modelscope/modelscope/pull/1378

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.27.1...v1.28.0

English Version

1. Set size_threshold_to_enforce_lfs to 1MB #1375
2. Limit MODELSCOPE_DOWNLOAD_PARALLELS to a maximum of 16 #1371
3. Fix some bugs in HubApi #1369

中文版本

1. 设置 size_threshold_to_enforce_lfs 为 1MB #1375
2. 将 MODELSCOPE_DOWNLOAD_PARALLELS 最大值限制为 16 #1371
3. 修复 HubApi 中的一些 bug #1369

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.27.0...v1.27.1

English Version

• Added an ASCII logo to the ModelScope command-line interface 😄.
• Increased the maximum number of supported files per repository to 50,000.
• Added requirements.txt to legacy model repositories to ensure backward compatibility, automatically installing required older versions of dependencies (transformers, huggingface_hub, diffusers) for proper model execution.
• Added ACoderPassBy/HifiSSR and ACoderPassBy/UnetVC model inference code with runnable README examples for voice cloning and speech enhancement tasks. #1208
• Enabled CLI upload to create repository directly if non-exists, streamlining model publishing workflow. #1351
• Fixed GPU memory leak issue in speaker_verification_res2net pipeline during multiple invocations. #1349
• Introduced scan_cache_dir utility to visualize local model cache status and optimize storage management. #1348
• Enhanced security framework with remote code execution verification mechanism to address vulnerability #1331, allowing explicit trust control for third-party model code via trust_remote_code parameter. #1333
• Optimized intra-cloud acceleration for model downloading with adaptive network protocol selection. #1315
• Implemented granular code security controls for importlib usage scenarios with precise file list filtering. #1323

中文版本

• ModelScope 命令行运行环境添加ASCII logo 😄。
• 放松单repo支持文件数到5w。
• 为保障一批早期模型正常运行，在这些模型仓库中新增requirements.txt文件，运行时将自动安装适配的旧版依赖库（transformers、huggingface_hub、diffusers等）以解决兼容性问题。
• 新增ACoderPassBy/HifiSSR和ACoderPassBy/UnetVC模型推理代码，提供可运行的README示例代码，支持语音克隆与语音增强任务 https://github.com/modelscope/modelscope/pull/1208
• 支持CLI命令行上传时自动创建目标仓库（若不存在），优化模型发布流程 https://github.com/modelscope/modelscope/pull/1351
• 修复speaker_verification_res2net流水线多次调用导致的显存泄露问题 https://github.com/modelscope/modelscope/pull/1349
• 新增本地缓存目录扫描功能（scan_cache_dir），可视化展示模型缓存状态并优化存储管理 https://github.com/modelscope/modelscope/pull/1348
• 修复安全漏洞（CVE-2023-1331），通过trust_remote_code参数实现第三方模型代码的分级信任控制 https://github.com/modelscope/modelscope/pull/1333
• 优化云上环境模型下载自适应全链路加速功能，支持智能网络协议选择 https://github.com/modelscope/modelscope/pull/1315
• 细化importlib使用场景的代码安全控制，实现模型下载过程的精准文件列表过滤 https://github.com/modelscope/modelscope/pull/1323

What's Changed

• Support trust_remote_code for pipeline and model by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1333
• Update publish.yaml ubuntu version by @Yunnglin in https://github.com/modelscope/modelscope/pull/1334
• Add setuptools for hub requirements by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1343
• [Feature] Add scan_cache_dir function by @Yunnglin in https://github.com/modelscope/modelscope/pull/1348
• Update 1.26.0 docker image dependency by @Yunnglin in https://github.com/modelscope/modelscope/pull/1339
• fix: 解决多次调用后的显存泄露。 by @biliyoubala in https://github.com/modelscope/modelscope/pull/1349
• remove master warning log by @yingdachen in https://github.com/modelscope/modelscope/pull/1336
• Allow upload to create repo (if not exists) directly by @yingdachen in https://github.com/modelscope/modelscope/pull/1351
• Fix typos in multiple files by @co63oc in https://github.com/modelscope/modelscope/pull/1330
• Fix typos by @co63oc in https://github.com/modelscope/modelscope/pull/1332
• vc ssr by @Z-yq in https://github.com/modelscope/modelscope/pull/1208
• fix: ollama template mapping by @suluyana in https://github.com/modelscope/modelscope/pull/1354
• Fix typos in multiple files by @co63oc in https://github.com/modelscope/modelscope/pull/1357

New Contributors

• @biliyoubala made their first contribution in https://github.com/modelscope/modelscope/pull/1349
• @Z-yq made their first contribution in https://github.com/modelscope/modelscope/pull/1208

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.26.0...v1.27.0

English Version

• Provided adaptive full-chain acceleration for model downloading in cloud environments. #1315
• Refined code security controls for importlib usage scenarios and supported more precise file list control during model downloading. #1323
• Added new Ollama templates, including cogito, mistral-small3.1, deepcoder, granite3.3, etc. #1318

中文版本

• 对于云上环境，提供自适应的模型下载全链路加速 https://github.com/modelscope/modelscope/pull/1315
• 对于importlib使用场景上细化的代码安全控制，并支持模型下载中更精准的文件列表控制 https://github.com/modelscope/modelscope/pull/1323
• 添加新的ollama template，包括cogito,mistral-small3.1,deepcoder,granite3.3等 https://github.com/modelscope/modelscope/pull/1318

What's Changed

• feat: new ollama template by @NononoSugar in https://github.com/modelscope/modelscope/pull/1318
• enable intra-cloud acceleration by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1315
• Support downloading exact file for hf wrapper by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1323
• Fix typos in multiple files by @co63oc in https://github.com/modelscope/modelscope/pull/1329
• Fix typos by @co63oc in https://github.com/modelscope/modelscope/pull/1328

New Contributors

• @NononoSugar made their first contribution in https://github.com/modelscope/modelscope/pull/1318

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.25.0...v1.26.0

V1.25.0 Release Note

English Version

1. Added hub APIs such as create_repo and delete_repo. https://github.com/modelscope/modelscope/pull/1282
2. Added swift mirrors. https://github.com/modelscope/modelscope/pull/1299

中文版本

1. 新增create_repo、delete_repo等hub API https://github.com/modelscope/modelscope/pull/1282
2. 新增swift镜像 https://github.com/modelscope/modelscope/pull/1299

What's Changed

• fix undefined name tf by @pengqu123 in https://github.com/modelscope/modelscope/pull/1294
• Add create_repo and upload UTs by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1282
• Set kwsbp==0.0.6 by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1293
• Add swift docker by @Jintao-Huang in https://github.com/modelscope/modelscope/pull/1299
• Merge release1.24 to master by @Jintao-Huang in https://github.com/modelscope/modelscope/pull/1298
• fix: Update speaker_verification_rdino_pipeline.py by @vra in https://github.com/modelscope/modelscope/pull/1300
• add more ignore patterns by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1303
• better error raise by @Jintao-Huang in https://github.com/modelscope/modelscope/pull/1301

New Contributors

• @pengqu123 made their first contribution in https://github.com/modelscope/modelscope/pull/1294
• @vra made their first contribution in https://github.com/modelscope/modelscope/pull/1300

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.24.1...v1.25.0

English Version

1. Add login for HFUtilTest #1288
2. Fix load dataset from huggingface hub #1275
3. Fix pipeline builder #1280
4. Lazy import hf modules #1283
5. Fix windows path #1286

中文版本

1. 为HFUtilTest添加登录功能 #1288
2. 修复从huggingface hub加载数据集 #1275
3. 修复pipeline构建器 #1280
4. 延迟导入hf模块 #1283
5. 修复windows路径 #1286

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.24.0...v1.24.1

English Version

1. Support python3.11 officially.
2. Add env variable MODELSCOPE_PREFER_AI_SITE to specify the endpoint to download models or datasets from. #1248
3. Add ollama templates: command-r7b-arabic、granite3.2-vision、phi4-mini、granite3.2、r1-1776、deepscaler、openthinker template #1247
4. Fix import issue #1253
5. Add snapshot_download headers for CI-TEST #1259
6. Fix external_engine_for_llm_checker #1260
7. Add more diffusers imports #1265
8. Fix downloading with specified class name #1270

中文版本

1. 支持python3.11
2. 添加环境变量MODELSCOPE_PREFER_AI_SITE，用于指定优先从哪个站点下载模型或数据集。#1248
3. 添加ollama模板：command-r7b-arabic、granite3.2-vision、phi4-mini、granite3.2、r1-1776、deepscaler、openthinker template #1247
4. 修复导入问题 #1253
5. 为CI-TEST添加snapshot_download headers #1259
6. 修复external_engine_for_llm_checker #1260
7. 添加更多的diffusers导入 #1265
8. 修复使用指定类名下载 #1270

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.23.2...v1.24.0

1. Hotfix commit info for uploading in the notebook env. #1244

English Version

1. [Breaking Change] Standardized the model download path from the default ~/.cache/modelscope/hub/<group-id>/<model-id> to ~/.cache/modelscope/hub/models/<group-id>/<model-id>. Additionally, in the official Docker image, the MODELSCOPE_CACHE has changed from /mnt/workspace/.cache/modelscope to /mnt/workspace/.cache/modelscope/hub. To prevent duplicate model downloads, we will automatically migrate models from the previous path to the new path.

2. The command line supports multiple filtering criteria for uploading and downloading models. Additionally, model uploads are pre-validated on the library side to prevent errors during the upload process.

3. Added Ollama support for models such as DeepSeek, olmo2, command-r7b, and others.

4. Improved support for automatically loading Hugging Face (HF) models and Pipelines. Currently, you can import various Auto components from ModelScope into Transformers, as well as PeftModel components from PEFT and Pipeline components from Diffusers. Added monkey patching for huggingface_hub and the from_pretrained-like method from transformers and peft to facilitate developers in using ModelScope to download and upload models. The patch for Hugging Face components is not enabled by default; developers need to manually enable it by using:

   from modelscope.utils.hf_util.patcher import patch_hub
   patch_hub()
   

   or a context version:

   from modelscope.utils.hf_util.patcher import patch_context
   with patch_context():
       # do something
   

5. [Break Change] Remove easyrobust from dependencies to upgrade the version of timm。If user want to use models related to easyrobust，please manullay install it by: pip install easyrobust

中文版本

1. [Break Change] 标准化了模型下载路径，从默认的~/.cache/modelscope/hub/<group-id>/<model-id>变为~/.cache/modelscope/hub/**models**/<group-id>/<model-id>，同时MODELSCOPE_CACHE在官方docker镜像中从/mnt/workspace/.cache/modelscope变为/mnt/workspace/.cache/modelscope/hub。为了防止模型重复下载，我们会自动将前版本路径的模型迁移到新的路径上

2. 命令行进行模型的上传/下载支持多重过滤条件。另外，模型上传在library侧进行提前校验，防止上传过程中出现错误

3. 增加了DeepSeek、olmo2、command-r7b等模型的ollama支持

4. 对HF的模型和Pipeline自动加载有了更好的支持。目前可以从modelscope中 import transformers的各类Auto组件，以及peft的PeftModel组件和diffusers的Pipeline组件。增加了对huggingface_hub和transformers和peft的from_pretrained等模型下载方法的monkey patch，方便开发者使用魔搭下载和上传模型。对huggingface组件的patch默认不会开启，开发者需要通过如下方式手动开启使用：

   from modelscope.utils.hf_util.patcher import patch_hub
   patch_hub()
   

   or a context version:

   from modelscope.utils.hf_util.patcher import patch_context
   with patch_context():
       # do something
   

5. [Break Change] 移除了easyrobust依赖，以便于升级timm依赖。如果使用easyrobust相关的模型，请手动安装该依赖：pip install easyrobust

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.22.3...v1.23.0

What's Changed

• standardize download directory for models/datasets by @yingdachen in https://github.com/modelscope/modelscope/pull/1197
• feat: new ollama template by @suluyana in https://github.com/modelscope/modelscope/pull/1199
• Add file size check in upload_folder by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1198
• clean directory structure by @yingdachen in https://github.com/modelscope/modelscope/pull/1201
• update upload progress by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1203
• remove MODELSCOPE_SHOW_INDIVIDUAL_PROGRESS_THRESHOLD by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1206
• Update upload func by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/1204
• Add more patches for hf by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1160
• clone and lint #1205 by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1209

English Version

1. Fix cookies issue due to cloud firewall token expired #1193
2. Fix file exists of sub folder #1192

中文版本

1. 修复云端防火墙token过期策略导致的本地cookies问题 #1193
2. 修复子文件夹存在文件的问题 #1192

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.22.2...v1.22.3

English Version

1. Add file name for uploading progress bar #1186
2. Fix commit failed issue for some cases
3. Fix upload login issue for notebook env.

中文版本

1. 为上传进度条添加文件名 #1186
2. 修复某些情况下的commit failed问题
3. 修复notebook环境下的上传登录问题

English Version

1. Support upload file and folder in the hub api #1152
2. Update doc with llama_index #1180
3. Unify datasets cache dir #1178
4. Fix/text gen #1177
5. Add repo_id and repo_type in snapshot_download #1172
6. Support all other ollama models #1174
7. Fix path concatenation to be windows compatible #1176
8. support ms-swift 3.0.0 #1166
9. Update downloading progress #1167
10. Fix iic/nlp_structbert_address-parsing_chinese_base does not support capitalize #1170
11. logger.warning when using remote code #1171
12. Support latest datasets #1163

中文版本

1. 支持使用CLI和hub api上传文件和文件夹 #1152
2. 优化llama_index的文档 #1180
3. 统一数据集缓存目录 #1178
4. 修复文本生成模型 #1177
5. 在snapshot_download中添加repo_id和repo_type #1172
6. 支持所有其他ollama模型 #1174
7. 修复路径拼接以支持windows #1176
8. 支持ms-swift 3.0.0 #1166
9. 更新下载进度条显示 #1167
10. 修复iic/nlp_structbert_address-parsing_chinese_base不支持capitalize #1170
11. 使用远程代码时添加logger.warning #1171
12. 支持最新的datasets #1163

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.22.0...v1.22.1

Hotfix

English Version

1. Fix command line for downloading datasets
2. Add thread_utils wrapper

中文版本

1. 修复数据集下载命令行

English Version

1. Optimize log recording #1080 #1081 #1093 #1089 #1099
2. Support launching LLM using the llamafile option via the command line #1087
3. Support automatic GPU usage for llamafile #1097
4. Better support for importing AutoClassfrom modelscope: #1098 #1106 #1107
5. The llm_first parameter in create_pipeline has been changed to external_engine_for_llm; the input for LLMPipeline must be in the messages format #1094
6. snapshot_download and dataset_snapshot_download support multi-threaded downloads, controlled by the max_workers parameter, with a default of 8 #1095 #1108
7. Support hash verification for downloaded files #1116

中文版本

1. 优化log记录 #1080 #1081 #1093 #1089 #1099
2. 支持命令行使用llamafile选项启动LLM #1087
3. 支持llamafile自动使用GPU #1097
4. 更完整的支持从modelscope导入AutoClass： #1098 #1106 #1107
5. create_pipeline参数llm_first变更为external_engine_for_llm ；LLMPipeline的input需为messages格式 #1094
6. snapshot_download和dataset_snapshot_download支持多线程下载，通过max_workers参数控制并发数量，默认数量为8 #1095 #1108
7. 支持下载文件的hash校验 #1116

What's Changed

• Update docker from release/1.20 by @Jintao-Huang in #1077
• Fix facial 68ldk detection by @Jintao-Huang in #1078
• Fix log for downloading to local dir by @yingdachen in #1080
• Log reduce by @yingdachen in #1081
• Fix dependency by @yingdachen in #1085
• Add llamafile support to command line by @yingdachen in #1087
• Refactor(models/audio/ans): Optimize zipformer and scaling layers by @Mashiro009 and @ZipEnhancer #1088
• Fix tensorflow warning by @Jintao-Huang in #1093
• Add AutoModelForImageSegmentation and T5EncoderModel by @yingdachen #1096
• Format llm pipeline by @yingdachen in #1094
• Lazy print ast logs by @Jintao-Huang in #1089
• Llamafile support gpu flag by @yingdachen in #1097
• Change warning to debug by @yingdachen in #1099
• More automodel by @yingdachen in #1098
• Add multi-thread download by @Yunnglin in #1095
• Fix potential double definition for OCR pipeline by @yingdachen in #1102
• Add transformer support for Qwen2vl by @yingdachen in #1106
• Add transformers compatibility for Vision2seq by @yingdachen in #1107
• Merge release 1.20 docker by @Jintao-Huang in #1109
• Support tag ci_image by @tastelikefeet in #1112
• Release transformers version to 4.33-4.46 by @Jintao-Huang in #1111
• Fix tqdm bar by @Yunnglin in #1108
• Handle unsupported Transformers class, and add more auto classes by @yingdachen in #1113
• Remove unnecessary code by @Jintao-Huang in #1115
• Add hash verification into cache file existence check by @yingdachen in #1116
• Fix accuracy case sensitiveness by @yingdachen in #1118
• Fix windows path by @yingdachen in #1119
• Make hash validation optional by @yingdachen in #1124
• Fix missing import by @yingdachen in #1126
• Skip obsolete sd pipeline by @tastelikefeet in #1131

New Contributors

• @Yunnglin made their first contribution in #1095

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.20.1...v1.21.0

Bug Fix:

1. Fix an import error which may cause snapshot_download fails when using a clean python env
2. Reduce the log of snapshot_download
3. Fix a bug which may cause the failure of facial_68ldk_detection pipeline

Bug修复：

1. 修复了一个在干净的python环境中可能引起snapshot_download依赖报错的问题
2. 减少snapshot_download的日志
3. 修复facial_68ldk_detection推理失败的问题

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.20.0...v1.20.1

English Version

1. New Models
   1. iic/speech_zipenhancer_ans_multiloss_16k_base. https://github.com/modelscope/modelscope/pull/1019
   2. AIDC-AI/Ovis1.6-Gemma2-9B. https://github.com/modelscope/modelscope/pull/1057
2. Hub Side:
   1. Created symbolic links in snapshot_download to avoid the issue of models with '.' in their names not being found. https://github.com/modelscope/modelscope/pull/1063
   2. Improved model upload by removing the requirement for configuration.json. https://github.com/modelscope/modelscope/pull/1062
   3. Added hub-api to check if the repository exists. https://github.com/modelscope/modelscope/pull/1060
3. Enhanced the template to_llama function to support more models. https://github.com/modelscope/modelscope/pull/1039, https://github.com/modelscope/modelscope/pull/1070.
4. Docker optimization and upgrades, removed unnecessary dependencies from the LLM image.

中文版本

1. 新模型
   1. iic/speech_zipenhancer_ans_multiloss_16k_base. https://github.com/modelscope/modelscope/pull/1019
   2. AIDC-AI/Ovis1.6-Gemma2-9B. https://github.com/modelscope/modelscope/pull/1057
2. Hub端：
   1. snapshot_download中创建软链接，避免模型名称中含'.'的模型无法被找到的问题。https://github.com/modelscope/modelscope/pull/1063
   2. 改善上传模型，移除对 configuration.json 的要求。https://github.com/modelscope/modelscope/pull/1062
   3. 添加仓库是否存在的hub-api。https://github.com/modelscope/modelscope/pull/1060
3. template to_llama功能增加更多支持的模型。https://github.com/modelscope/modelscope/pull/1039, https://github.com/modelscope/modelscope/pull/1070
4. docker优化与升级, llm镜像移除不需要的依赖。

What's Changed

• Fix timestamp in docker build by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1049
• feat(audio/ans): Add ZipEnhancer and related layers for acoustic nois… by @Mashiro009 in https://github.com/modelscope/modelscope/pull/1019
• Fix the slow downloading by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1051
• Fix bash and transformers version by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1053
• Fix some bugs by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1056
• fix(audio ans pipeline): Restore file reading from string input in ANSZip… by @Mashiro009 in https://github.com/modelscope/modelscope/pull/1055
• OCR pipeline shall depend on TF only when necessary by @yingdachen in https://github.com/modelscope/modelscope/pull/1059
• fix: text error correction batch run bug by @smartmark-pro in https://github.com/modelscope/modelscope/pull/1052
• add log for download location by @yingdachen in https://github.com/modelscope/modelscope/pull/1061
• add repo existence check hub-api by @yingdachen in https://github.com/modelscope/modelscope/pull/1060
• improve upload model, remove requirment for configuration.json by @yingdachen in https://github.com/modelscope/modelscope/pull/1062
• Template.to_ollama: add new argument split by @suluyana in https://github.com/modelscope/modelscope/pull/1039
• Feat(multimodal model):ovis vl pipeline by @suluyana in https://github.com/modelscope/modelscope/pull/1057
• default install tf-keras by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1064
• Symbolic link by @yingdachen in https://github.com/modelscope/modelscope/pull/1063
• Fix the missing init.py file. by @Jintao-Huang in https://github.com/modelscope/modelscope/pull/1066
• try to reduce the image size of llm by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1067
• fix numpy build error by @tastelikefeet in https://github.com/modelscope/modelscope/pull/1068
• fix docker numpy version by @Jintao-Huang in https://github.com/modelscope/modelscope/pull/1069
• feat ollama template: llama3.2-vision by @suluyana in https://github.com/modelscope/modelscope/pull/1070
• update docker evalscope version by @Jintao-Huang in https://github.com/modelscope/modelscope/pull/1071
• update docker by @Jintao-Huang in https://github.com/modelscope/modelscope/pull/1073
• update docker by @Jintao-Huang in https://github.com/modelscope/modelscope/pull/1075
• Update llm docker by @Jintao-Huang in https://github.com/modelscope/modelscope/pull/1076

New Contributors

• @Mashiro009 made their first contribution in https://github.com/modelscope/modelscope/pull/1019
• @smartmark-pro made their first contribution in https://github.com/modelscope/modelscope/pull/1052

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.19.2...v1.20.0

Hotfix: Set datasets<=3.0.1 to fix datasets import error

English Version

1. Update version of outlines and vllm in docker file： https://github.com/modelscope/modelscope/pull/1034
2. Fix processor and feature extractor in hf_util： https://github.com/modelscope/modelscope/pull/1031
3. Fix device_map issue for gptq： https://github.com/modelscope/modelscope/pull/1027
4. Fix Lint issue：https://github.com/modelscope/modelscope/pull/1025
5. Fix if-else condition： https://github.com/modelscope/modelscope/pull/1024

中文版本

1. 增加dockerfile中outlines版本限制和更新vllm版本限制： https://github.com/modelscope/modelscope/pull/1034
2. 修复hf_util中processor和feature extractor： https://github.com/modelscope/modelscope/pull/1031
3. 修复gptq使用中的device_map问题： https://github.com/modelscope/modelscope/pull/1027
4. 修复commit格式：https://github.com/modelscope/modelscope/pull/1025
5. 修复判断条件： https://github.com/modelscope/modelscope/pull/1024

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.19.0...release/1.19

English Version

1. Add clear-cache for modelscope command line , refer to: https://github.com/modelscope/modelscope/pull/1009
2. Adapt datasets>=3.0，refer to: https://github.com/modelscope/modelscope/pull/1002
3. Add template module: https://github.com/modelscope/modelscope/pull/995
4. Ignore pt/pth format files for AutoTokenizer: https://github.com/modelscope/modelscope/pull/1008
5. Unify the log format in MsDataset: https://github.com/modelscope/modelscope/pull/997
6. Improve the coverage of loading&previewing datasets from modelscope hub.

中文版本

1. modelscope command line, clear-cache功能，参考： https://github.com/modelscope/modelscope/pull/1009
2. datasets适配最新的3.0.x版本，参考：https://github.com/modelscope/modelscope/pull/1002
3. 支持template（将swift中的template模块迁移到modelscope lib）： https://github.com/modelscope/modelscope/pull/995
4. tokenizer加载时忽略以下格式文件：.pt,.pth: https://github.com/modelscope/modelscope/pull/1008
5. 统一dataset 下载log： https://github.com/modelscope/modelscope/pull/997
6. 提升数据集加载和预览的覆盖率

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.18.1...release/1.19

English Version

1. Fix a bug that zero-sized file cannot be download
2. Support the hub patching to fit the latest version of vllm or other frameworks

中文版本

1. 修复了零大小文件无法下载的问题
2. 支持对modelhub的patch，以适配最新的vllm或其他的开源代码库的魔搭模型下载功能

What's Changed

• use tqdm auto by @yingdachen in https://github.com/modelscope/modelscope/pull/982
• Support create file with size 0 by @tastelikefeet in https://github.com/modelscope/modelscope/pull/984
• patch hf hub by @tastelikefeet in https://github.com/modelscope/modelscope/pull/987
• Refactor zero sized file downloading by @tastelikefeet in https://github.com/modelscope/modelscope/pull/991

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.18.0...v1.18.1

English Version

Dataset features

• Load repo files downloaded by modelscope download --dataset with MsDataset.load

New models

No.	Model-id and links
1	Facial 68 landmark detection

Chinese Version

数据集

• 对于使用modelscope命令行modelscope download --dataset下载的数据集，支持使用MsDataset.load加载

新模型

No.	Model-id and links
1	真人人脸68个关键点检测

What's Changed

• Fix page in list_repo by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/938
• Release/1.17 by @suluyana in https://github.com/modelscope/modelscope/pull/942
• fix file download off by one bug by @liuyhwangyh in https://github.com/modelscope/modelscope/pull/943
• fix compatible of origin command line by @liuyhwangyh in https://github.com/modelscope/modelscope/pull/944
• fix docs build by @yingdachen in https://github.com/modelscope/modelscope/pull/947
• Fix streaming load issues by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/953
• Fix datasets issues by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/948
• [fix #934] allow logging level to be configurable by @yingdachen in https://github.com/modelscope/modelscope/pull/958
• log download file name by @yingdachen in https://github.com/modelscope/modelscope/pull/965
• Fix patching hf by @tastelikefeet in https://github.com/modelscope/modelscope/pull/969
• add facial 68 landmark detection model by @gylcalex1314 in https://github.com/modelscope/modelscope/pull/964
• lint code by @tastelikefeet in https://github.com/modelscope/modelscope/pull/970

New Contributors

• @yingdachen made their first contribution in https://github.com/modelscope/modelscope/pull/947
• @gylcalex1314 made their first contribution in https://github.com/modelscope/modelscope/pull/964

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.17.0...v1.18.0

New models

No.	Model-id and links
1	单目人体图像法向估计

Highlight features

1. Included the --token parameter in the command line interface.
2. The snapshot_download function can be specified with parameters ignore_patterns and allowed_patterns.

What's Changed

1. Fix page in list_repo 938
2. Fixed the pagination bug in the dataset file download function. Integrated transformers into the framework. Added Pillow to the framework. 927
3. Update README_zh.md 928

New models

No.	Model-id and links
1	ERes2NetV2说话人确认-中文-通用-200k-Spkrs

Highlight features

1. Refactor AST index file generator
2. Support swift model inference for llm pipeline
3. Refactor installation dependencies

What's Changed

• Add Eres2netv2ep4 https://github.com/modelscope/modelscope/pull/892
• Refactor AST index file generator https://github.com/modelscope/modelscope/pull/897
• Support swift model inference for llm pipeline https://github.com/modelscope/modelscope/pull/880
• Refactor installation dependencies https://github.com/modelscope/modelscope/pull/889
• Shorten check model timeout and snapshot_download file https://github.com/modelscope/modelscope/pull/896
• exchange the order of replacement https://github.com/modelscope/modelscope/pull/893
• add command line usage, optimize command line log https://github.com/modelscope/modelscope/pull/888

New models

No.	Model-id and links
1	TDNN说话人确认-英文-VoxCeleb-16k-离线-pytorch
2	SDPN说话人确认-英文-VoxCeleb-16k-离线-pytorch
3	glm-4-9b series

Highlight features

1. Supports resumption of downloads from breakpoints.
2. optimized download progress bar better display granularity, better experience under low bandwidth.
3. add login and download command tool.

What's Changed

• add donwload command line and local_dir parameter @liuyhwangyh in https://github.com/modelscope/modelscope/pull/866
• Download optimize @liuyhwangyh in https://github.com/modelscope/modelscope/pull/862
• add sdpn and tdnn @yfchenmodelscope in https://github.com/modelscope/modelscope/pull/865
• fix issue #845 @liuyhwangyh in https://github.com/modelscope/modelscope/pull/861
• BUGFIX: TypeError: exceptions must derive from BaseException by @AlphaHinex in https://github.com/modelscope/modelscope/pull/859
• add eres2netv2 by @yfchenmodelscope in https://github.com/modelscope/modelscope/pull/830
• Fix json.decoder.JSONDecodeError when load pipeline_schema.json @ in https://github.com/modelscope/modelscope/pull/810
• remove necessary dependency of transformer @wenmengzhou in https://github.com/modelscope/modelscope/pull/857
• update readme with latest model and github trending badage by @wenmengzhou in https://github.com/modelscope/modelscope/pull/851

New Contributors

• @AlphaHinex made their first contribution in https://github.com/modelscope/modelscope/pull/859

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.14.0...v1.15.0

New models

No.	Model-id and links
1	Qwen1.5-110B-Chat
2	CodeQwen1.5-7B-Chat
3	WizardLM-2-8x22B
4	c4ai-command-r-v01
5	通义千问1.5-32B-对话
6	dbrx-instruct

Highlight features

1. Dataset refactoring, to be compatible with HF datasets structure, new to limit datasets<2.19.0 (Breaking Changes)

What's Changed

• Fix api get_dataset_file_url by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/811
• Feature/raft by @pipikk in https://github.com/modelscope/modelscope/pull/702
• fix installation error in README_zh.md by @cyx2000 in https://github.com/modelscope/modelscope/pull/821
• BUGFIX: TypeError: exceptions must derive from BaseException by @joeyism in https://github.com/modelscope/modelscope/pull/816
• add eres2netv2 by @yfchenmodelscope in https://github.com/modelscope/modelscope/pull/830
• add input output for siamese by @WeiLai5432 in https://github.com/modelscope/modelscope/pull/810
• Fix datasets hf utils by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/834
• merge from release 1.13.3 by @zzhangpurdue in https://github.com/modelscope/modelscope/pull/837

New Contributors

• @pipikk made their first contribution in https://github.com/modelscope/modelscope/pull/702
• @cyx2000 made their first contribution in https://github.com/modelscope/modelscope/pull/821
• @joeyism made their first contribution in https://github.com/modelscope/modelscope/pull/816
• @WeiLai5432 made their first contribution in https://github.com/modelscope/modelscope/pull/810

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.13.2...v1.14.0

Highlight features

1. Dataset refactoring, to be compatible with HF datasets structure. (Breaking Changes)
2. Unified datasets storage and management with GIT. (Breaking Changes)

What's Changed

• upload marigold monocular depth estimation core files by @Ranqing in https://github.com/modelscope/modelscope/pull/703
• chore: Formal LICENSE content by @tisonkun in https://github.com/modelscope/modelscope/pull/799
• add ViViT-demo by @ccyhxg in https://github.com/modelscope/modelscope/pull/796
• move doc to classroom by @tastelikefeet in https://github.com/modelscope/modelscope/pull/802
• Fix error "modelscope attributeerror: 'dict' object has no attribute … by @wertycn in https://github.com/modelscope/modelscope/pull/800
• fix download file with spical name as 'Image+Title.png' by @liuyhwangyh in https://github.com/modelscope/modelscope/pull/805
• Dataset refactor by @wangxingjun778 in https://github.com/modelscope/modelscope/pull/807

New Contributors

• @Ranqing made their first contribution in https://github.com/modelscope/modelscope/pull/703
• @tisonkun made their first contribution in https://github.com/modelscope/modelscope/pull/799
• @wertycn made their first contribution in https://github.com/modelscope/modelscope/pull/800

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.13.1...v1.13.2

New models

No.	Model-id and links
1	GeoMVSNet:基于几何感知的多视图深度估计
2	Res2Net说话人确认-中文-3D-Speaker-16k
3	ResNet34说话人确认-中文-3D-Speaker-16k
4	自监督深度补全

Highlight features

1. Support importing AWQConfig from modelscope
2. Support stream_generate for LLMPipeline

What's Changed

• UViT ImageNet by @ccyhxg in https://github.com/modelscope/modelscope/pull/763
• Add DiT_ImageNet_Demo by @ccyhxg in https://github.com/modelscope/modelscope/pull/767
• Add a QA example based on llamaindex (with qwen1.5) by @yhxx511 in https://github.com/modelscope/modelscope/pull/759
• Fix words by @co63oc in https://github.com/modelscope/modelscope/pull/747
• Fix word pubicly by @co63oc in https://github.com/modelscope/modelscope/pull/748
• Fix word pipleline by @co63oc in https://github.com/modelscope/modelscope/pull/749
• Fix word orignal by @co63oc in https://github.com/modelscope/modelscope/pull/750
• add awqconfig by @Jintao-Huang in https://github.com/modelscope/modelscope/pull/761
• add SiT_ImageNet_Demo by @ccyhxg in https://github.com/modelscope/modelscope/pull/770
• Fix words StableDiffuisonExporter -> StableDiffusionExporter by @co63oc in https://github.com/modelscope/modelscope/pull/746
• add self supervised depth completion. by @heyyxd in https://github.com/modelscope/modelscope/pull/711
• To solve the "ImportError: always import a name 'LlamaTokenizer' from… by @Siu-Ming in https://github.com/modelscope/modelscope/pull/745
• Fix code of conduct by @tastelikefeet in https://github.com/modelscope/modelscope/pull/774
• add res2net resnet models by @yfchenmodelscope in https://github.com/modelscope/modelscope/pull/772
• upgrade image build by @liuyhwangyh in https://github.com/modelscope/modelscope/pull/773
• change output video format from mp4v to mp4 h264 by @qslia in https://github.com/modelscope/modelscope/pull/757
• fix text_to_video_synthesis_model device by @slin000111 in https://github.com/modelscope/modelscope/pull/751
• add qwen1.5_doc_search_QA_based_on_langchain.ipynb by @ccyhxg in https://github.com/modelscope/modelscope/pull/769
• Add files via upload by @ccyhxg in https://github.com/modelscope/modelscope/pull/788
• pre-commit passed by @tastelikefeet in https://github.com/modelscope/modelscope/pull/789
• fix build issue by @liuyhwangyh in https://github.com/modelscope/modelscope/pull/786
• fix SeqGPTPipeline input force cuda issue by @RainJayTsai in https://github.com/modelscope/modelscope/pull/738
• fix image_portrait_enhancement_pipeline rgb channel issue by @sun11 in https://github.com/modelscope/modelscope/pull/740
• remove download interval check by @liuyhwangyh in https://github.com/modelscope/modelscope/pull/771
• Fix initilize initialize, etc by @co63oc in https://github.com/modelscope/modelscope/pull/781
• fix pre-commit by @tastelikefeet in https://github.com/modelscope/modelscope/pull/794
• Features/cv_geomvsnet_multi_view_depth_estimation_general by @shengzhesz in https://github.com/modelscope/modelscope/pull/790
• Support stream_generate for LLMPipeline by @Firmament-cyou in https://github.com/modelscope/modelscope/pull/768

New Contributors

• @heyyxd made their first contribution in https://github.com/modelscope/modelscope/pull/711
• @Siu-Ming made their first contribution in https://github.com/modelscope/modelscope/pull/745
• @yfchenmodelscope made their first contribution in https://github.com/modelscope/modelscope/pull/772
• @qslia made their first contribution in https://github.com/modelscope/modelscope/pull/757
• @RainJayTsai made their first contribution in https://github.com/modelscope/modelscope/pull/738
• @sun11 made their first contribution in https://github.com/modelscope/modelscope/pull/740
• @shengzhesz made their first contribution in https://github.com/modelscope/modelscope/pull/790

Full Changelog: https://github.com/modelscope/modelscope/compare/v1.12.0...v1.13.1