Mirrors/modelscope
1
0
Fork 0
mirror of https://github.com/modelscope/modelscope.git synced 2026-02-24 04:01:10 +01:00
Code Issues Packages Projects Releases Wiki Activity
2,794 Commits 267 Branches 107 Tags
dev/merge_release128
Commit Graph

2794 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
xingjun.wxj
d8f0e39d1e merge release/1.28 2025-08-06 17:05:09 +08:00
xingjun.wxj
f5868126b4 ok Merge branch 'release/1.28' of github.com:modelscope/modelscope into dev/merge_release128 2025-08-06 17:04:20 +08:00
Koko-ry
1d93f7e56d Feat: Add AIGC model creation support to create_model (#1431) 
1.  Add create aigc model in `create_model` func
2. Add CLI for creating aigc model
 2025-08-06 17:01:42 +08:00
suluyana
53ceca4df4 feat: sentence_embedding pipeline (#1435) 2025-08-06 15:43:36 +08:00
xingjun.wxj
2f792d936f set version to 1.28.2 v1.28.2 2025-08-06 15:38:06 +08:00
Xingjun.Wang
341857cc8d fix features for datasets<=3.6.0 (#1447) 2025-08-06 15:36:24 +08:00
Xingjun.Wang
ee05e12d75 Add create repo cli (#1445) 2025-08-05 21:42:41 +08:00
Koko-ry
51e28f5a49 Feat: add atomic capabilities - MCPApi (#1426) 2025-08-05 21:41:26 +08:00
tastelikefeet
d8bb003681 fix bug (#1441) 2025-08-04 22:05:49 +08:00
Xingjun.Wang
7005876f84 add revision arg for upload_file (#1440) 2025-08-04 21:42:00 +08:00
Yunlin Mao
ff7551c494 fix large log in scan cache test (#1439) 2025-08-04 20:58:30 +08:00
tastelikefeet
3bad15b779 add ms-agent into dockerfile (#1436) 2025-08-04 17:08:48 +08:00
suluyana
b3d24d18e6 ollama template: qwen3-coder, gemma3n, mistral-small3.2, magistral, devstral (#1434) 
Co-authored-by: suluyan <suluyan.sly@aliabab-inc.com>
 2025-08-04 17:07:06 +08:00
jinghanhu
b537efebed Merge release 1.28 (#1401) 2025-08-04 17:00:14 +08:00
Tian-Jionglu
7ed173be8b fix(deps): add missing filelock dependency (#1427) 
Co-authored-by: Tian-Jionglu <tianjionglu@outlook.com>
 2025-08-04 16:57:45 +08:00
tastelikefeet
556aa76dc3 fix pyyaml according to: https://github.com/Anchor0221/CVE-2025-50460 (#1428) 2025-08-01 22:04:44 +08:00
xingjun.wxj
595f3ea263 merge master and set version to 1.28.1 v1.28.1 2025-07-25 14:36:26 +08:00
xingjun.wxj
4854f1e5e5 ok Merge branch 'master' of github.com:modelscope/modelscope into release/1.28 2025-07-25 14:35:06 +08:00
Xingjun.Wang
49d50b2126 fix use_cookies in get_model_files (#1424) 2025-07-25 14:24:27 +08:00
Xingjun.Wang
e173fd9619 Add MODELSCOPE_CREDENTIALS_PATH for multi-users env (#1423) 2025-07-22 17:45:28 +08:00
Xingjun.Wang
62078f1796 Fix dataset infos (#1414) 2025-07-22 17:45:19 +08:00
Koko-ry
f3d014d3ce Skip UT TextToSpeechSambertHifigan16kPipelineTest temporarily (#1420) 2025-07-22 17:45:07 +08:00
tastelikefeet
d031f3e20b weak file lock (#1417) 2025-07-22 17:44:57 +08:00
Koko-ry
3d11b891ca Fix RCE issue for plugins (#1415) 
1. Add `trust_remote_code` in func `from_pretrained`  with plugins application
 2025-07-22 17:44:48 +08:00
Xingjun.Wang
2b3747db53 Add MODELSCOPE_CREDENTIALS_PATH for multi-users env (#1423) 2025-07-22 14:22:00 +08:00
Xingjun.Wang
e97974d1c0 Fix dataset infos (#1414) 2025-07-17 19:27:39 +08:00
Koko-ry
d6f11c1c22 Skip UT TextToSpeechSambertHifigan16kPipelineTest temporarily (#1420) 2025-07-17 18:57:44 +08:00
tastelikefeet
f01864a2f9 weak file lock (#1417) 2025-07-17 18:38:24 +08:00
Koko-ry
1d24435275 Fix RCE issue for plugins (#1415) 
1. Add `trust_remote_code` in func `from_pretrained`  with plugins application
 2025-07-16 14:33:31 +08:00
hjh0119
c63952e90c fix vaehook custom_group_norm shape dismatch v1.28.0 2025-07-08 21:14:31 +08:00
hjh0119
339b06ca41 revert flash-attn dep 2025-07-07 17:07:44 +08:00
hjh0119
b8ff6abc66 Merge remote-tracking branch 'upstream/master' into release/1.28 2025-07-07 14:24:35 +08:00
tastelikefeet
6c619155ac Fix processing ignore files (#1398) 2025-07-07 14:22:33 +08:00
hjh0119
48c777f4c3 revert vllm&torch dep 2025-07-07 14:01:51 +08:00
hjh0119
5a8f97cbd1 update docker image for 1.28 2025-07-07 10:34:30 +08:00
hjh0119
ac7404ce87 bump version 2025-07-04 11:35:38 +08:00
Xingjun.Wang
f4217e6183 Upgrade datasets (#1393) 
* upgrade datasets to 3.6.0

* fix mcp lint
 2025-07-04 10:59:07 +08:00
Xingjun.Wang
88406c17ad Add upload check envs (#1389) 
* add SIZE_THRESHOLD_TO_ENFORCE_LFS

* Add upload check envs
 2025-07-04 10:28:54 +08:00
MengAiDev
f023b2514c perf(vae): optimize tiling performance and memory usage (#1378) 
* perf(vae): optimize tiling performance and memory usage

- Improve memory management by moving tensors to CPU when not in use
- Enhance GroupNormParam initialization and accumulation logic
- Simplify task queue processing and improve NaN handling
- Optimize tile preparation and result assembly
- Add torch.cuda.empty_cache() calls to free up GPU memory

* remove unused import

* lint

* fix cuda empty cache and dtype

* lint pass

---------

Co-authored-by: hjh0119 <hujinghan.hjh@alibaba-inc.com>
 2025-07-04 10:28:25 +08:00
jinghanhu
57d88c06e7 Merge from release 1.27 #1392 2025-07-03 20:10:39 +08:00
Koko-ry
bc58eac935 Feat deletefiles (#1384) 
1. Add delete files
2. Add mcp api based on OpenAPI
 2025-07-03 19:51:22 +08:00
suluyan
7230561cbd image deps upgrade: transformers, huggingface_hub, node.js 2025-07-03 15:59:20 +08:00
suluyan
6a895360da fix boost.tar.gz not in gzip format 2025-07-03 15:59:12 +08:00
hjh0119
dee7d0576d fix pgp fail 2025-07-03 15:58:10 +08:00
Xingjun.Wang
7c9b89d24d add SIZE_THRESHOLD_TO_ENFORCE_LFS (#1383) 2025-06-26 14:11:27 +08:00
Jintao
29ea57dfbc [hub] Support ProgressCallback (#1380) 
1. Add ProgressCallback for file download
 2025-06-26 14:02:05 +08:00
Xingjun.Wang
694fc89735 Set size_threshold_to_enforce_lfs to 1MB (#1375) 
* Set the maximum value of MODELSCOPE_DOWNLOAD_PARALLELS to 16

* set size_threshold_to_enforce_lfs to 1MB
 2025-06-19 17:06:29 +08:00
Xingjun.Wang
d092a4350c Set the maximum value of MODELSCOPE_DOWNLOAD_PARALLELS to 16 (#1371) 2025-06-19 15:49:31 +08:00
Xingjun.Wang
c9d0c46412 fix some bugs in HubApi (#1369) 2025-06-19 10:36:44 +08:00
tastelikefeet
86dd6c3434 fix some transformers version missing the chat template definition (#1368) 2025-06-11 20:16:24 +08:00