[high risk]Speedup 2.5-6X moe tuning and pin ngram to cpu #2303
update
5fe714d6
[pre-commit.ci] auto fixes from pre-commit.com hooks
b0556ad4
update
9f9331c3
update
f644f237
Merge branch 'main' into fix_moe_loading
fd6f5f04
[pre-commit.ci] auto fixes from pre-commit.com hooks
c0b2cee2
disable all moe replacements
84fe3554
Merge branch 'fix_moe_loading' of https://github.com/intel/auto-round…
481f1a3f
speedup tuning
afd47ffd
no speedup gain
8d0edbf6
[pre-commit.ci] auto fixes from pre-commit.com hooks
1b36314b
update
a1fbeada
Merge branch 'suport_fuse_moe' of https://github.com/intel/auto-round…
640e31b3
update
a68ad421
better support multi-gpus
1524006d
wenhuach21
changed the title Speedup 3X moe tuning Speedup 3X moe tuning via grouped experts 16 days ago
Merge branch 'main' into suport_fuse_moe
1337a486
[pre-commit.ci] auto fixes from pre-commit.com hooks
a1a5b4d8
clean
ce8be474
Merge branch 'main' into suport_fuse_moe
47254b8d
update
a7ee60b8
[pre-commit.ci] auto fixes from pre-commit.com hooks
d98505dd
update
8f0af611
Merge branch 'suport_fuse_moe' of https://github.com/intel/auto-round…
37573f8a
support group_mm in monkey patch
e5e67e17
[pre-commit.ci] auto fixes from pre-commit.com hooks
610b9239
fix qwen4
28c72d98
Merge branch 'suport_fuse_moe' of https://github.com/intel/auto-round…
b0757577
try to fix ngram issue
579f4491
fix
052b5cdd
try to fix some glm oom issue
5991307e
[pre-commit.ci] auto fixes from pre-commit.com hooks
3f489391
try to fix glm oom issue
52f8b3c3
Merge branch 'suport_fuse_moe' of https://github.com/intel/auto-round…
db91535d
try to fix glm oom issue
63002e1a
[pre-commit.ci] auto fixes from pre-commit.com hooks
557d5d4f
fix
6d2f7699
Merge branch 'suport_fuse_moe' of https://github.com/intel/auto-round…
2ffed45d
Merge branch 'main' into suport_fuse_moe
685332b4
fix
30ad3754
wenhuach21
changed the title Speedup 3X moe tuning via grouped experts [high risk]Speedup 2.5-4X moe tuning via grouped experts and pin ngram to cpu to fix qwen ram issue 12 days ago
Merge branch 'suport_fuse_moe' of https://github.com/intel/auto-round…
c0cddc54
clean
6a97f983
wenhuach21
added this to the 0.16.0 milestone 12 days ago
Potential fix for pull request finding
e35bcd49
fix issue
f960c8ee
Merge branch 'suport_fuse_moe' of https://github.com/intel/auto-round…
96c067ad
[pre-commit.ci] auto fixes from pre-commit.com hooks
14f3a2e5
wenhuach21
changed the title [high risk]Speedup 2.5-4X moe tuning via grouped experts and pin ngram to cpu to fix qwen ram issue [high risk]Speedup 2.5-4X moe tuning and pin ngram to cpu 12 days ago
wenhuach21
changed the title [high risk]Speedup 2.5-4X moe tuning and pin ngram to cpu [high risk]Speedup 2.5-6X moe tuning and pin ngram to cpu 12 days ago
update log
a63adbc3
Merge branch 'suport_fuse_moe' of https://github.com/intel/auto-round…
ffb592af
[pre-commit.ci] auto fixes from pre-commit.com hooks
7578ec0a
update log
38868ffd
Merge branch 'suport_fuse_moe' of https://github.com/intel/auto-round…
a2c2317d
try to fix ut
96930d58
try to fix ut
e334b63c
wenhuach21
deleted the suport_fuse_moe branch 11 days ago
Assignees
No one assigned
Login to write a write a comment.
Login via GitHub