auto-round
Support for immediate saving to reduce ram usage
#965
Merged

Support for immediate saving to reduce ram usage #965

wenhuach21 merged 38 commits into main from kaihui/save_block
Kaihui-intel
Kaihui-intel save per block
9da69a88
Kaihui-intel enable multi block save
dc24682b
Kaihui-intel support export save
856ab066
Kaihui-intel update rtn support
97cbfdbd
Kaihui-intel support
fb026aa6
Kaihui-intel rebase main
bb6113f9
Kaihui-intel del utils.py
148f085d
pre-commit-ci[bot] [pre-commit.ci] auto fixes from pre-commit.com hooks
afb6eefb
wenhuach21
wenhuach21 wenhuach21 requested a review from n1ck-guo n1ck-guo 230 days ago
wenhuach21 wenhuach21 requested a review from yiliu30 yiliu30 230 days ago
wenhuach21 wenhuach21 requested a review from xin3he xin3he 230 days ago
wenhuach21
wenhuach21 commented on 2025-10-30
Kaihui-intel fix args
b2c14dfc
xin3he xin3he added this to the 1.0 milestone 229 days ago
xin3he xin3he removed this from to the 1.0 milestone 229 days ago
xin3he xin3he added this to the 0.9.0 milestone 229 days ago
Kaihui-intel optimize memory and fix rtn module
38297788
Kaihui-intel revert max_shard_size
8f388395
Kaihui-intel move save_block_immediate into utils
70d873e2
pre-commit-ci[bot] [pre-commit.ci] auto fixes from pre-commit.com hooks
f4ace2cd
pre-commit-ci[bot] [pre-commit.ci] auto fixes from pre-commit.com hooks
c3142310
Kaihui-intel fix import
f64cfcf0
Kaihui-intel update max_shard_size
e471148e
Kaihui-intel Merge branch 'kaihui/save_block' of https://github.com/intel/auto-rou…
b19db96f
Kaihui-intel Kaihui-intel requested a review from wenhuach21 wenhuach21 229 days ago
wenhuach21 wenhuach21 changed the title Support for immediate saving [High Risk]Support for immediate saving 229 days ago
wenhuach21 wenhuach21 requested a review from WeiweiZhang1 WeiweiZhang1 229 days ago
wenhuach21 wenhuach21 requested a review from hshen14 hshen14 229 days ago
wenhuach21
wenhuach21 commented on 2025-10-31
wenhuach21
wenhuach21 commented on 2025-10-31
wenhuach21
wenhuach21 commented on 2025-10-31
wenhuach21
wenhuach21 approved these changes on 2025-10-31
Kaihui-intel remove is_meta_model
02f891d6
pre-commit-ci[bot] [pre-commit.ci] auto fixes from pre-commit.com hooks
f6671ed3
Kaihui-intel merge main
c1da5e56
Kaihui-intel add uts
39b54705
pre-commit-ci[bot] [pre-commit.ci] auto fixes from pre-commit.com hooks
579a7f80
Kaihui-intel fix gpu ut model_name
f1793e05
wenhuach21
wenhuach21 requested changes on 2025-10-31
Kaihui-intel set immediate packing saving to True
5aa3f1ef
pre-commit-ci[bot] [pre-commit.ci] auto fixes from pre-commit.com hooks
67e2591f
Kaihui-intel Merge branch 'main' into kaihui/save_block
ca55a906
Kaihui-intel flow saving setting
7a21c5aa
Kaihui-intel check gguf
e5fd4f24
Kaihui-intel pack layer_names immediately
5367af64
wenhuach21
wenhuach21 commented on 2025-11-05
wenhuach21
wenhuach21 commented on 2025-11-05
wenhuach21
wenhuach21 commented on 2025-11-05
Kaihui-intel wrapper _immediate_pack & rm uts & pop lm_head & rm expose args
5065336b
Kaihui-intel revert ut
ceb2a92d
pre-commit-ci[bot] [pre-commit.ci] auto fixes from pre-commit.com hooks
5375be6d
wenhuach21
Kaihui-intel add low_cpu_mem_usage
70f59ec3
Kaihui-intel merge main
bd1532b4
pre-commit-ci[bot] [pre-commit.ci] auto fixes from pre-commit.com hooks
a551e131
Kaihui-intel fix conflict
50a82e2f
wenhuach21
wenhuach21 commented on 2025-11-07
wenhuach21
wenhuach21 commented on 2025-11-07
wenhuach21
wenhuach21 commented on 2025-11-07
wenhuach21
wenhuach21 commented on 2025-11-07
Kaihui-intel update low_cpu_mem_usage
3c682c65
wenhuach21
wenhuach21 commented on 2025-11-07
wenhuach21
wenhuach21 commented on 2025-11-07
Kaihui-intel move to last arg
35750399
wenhuach21 wenhuach21 requested a review from wenhuach21 wenhuach21 222 days ago
wenhuach21 wenhuach21 changed the title [High Risk]Support for immediate saving Support for immediate saving to reduce ram usage 222 days ago
wenhuach21
wenhuach21 approved these changes on 2025-11-07
wenhuach21 wenhuach21 merged daeb3bb7 into main 222 days ago
wenhuach21 wenhuach21 deleted the kaihui/save_block branch 222 days ago

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone