PR #18175 llama 70b model fusion and shardding

llama 70b model fusion and shardding #18175

frank-dong-ms merged 15 commits into microsoft:main from frank-dong-ms:frdong/shard-llama

add shardding support for llama model and update convert/benchmark sc…

c958bc6e

github-advanced-security commented on 2023-10-30

fix bug in llama input

dd41176c

frank-dong-ms requested a review from

kunal-vaishnavi 2 years ago

frank-dong-ms requested a review from

tianleiwu 2 years ago

frank-dong-ms added release:1.16.2

frank-dong-ms added sdxl_llama

tianleiwu commented on 2023-10-31

kunal-vaishnavi commented on 2023-10-31

lint style fix

92066936

take comments

b30ff30b

github-advanced-security commented on 2023-10-31

handle export device separately for different model

c4649e8a

github-advanced-security commented on 2023-10-31

lint fix

f393cb2a

kunal-vaishnavi commented on 2023-10-31

Merge branch 'microsoft:main' into frdong/shard-llama

59a27e2e

mindest commented on 2023-10-31

take comments

fdd4c5fd

github-advanced-security commented on 2023-10-31

take comments

79e185dc

fix comments

0093725a

hanbitmyths commented on 2023-11-01

kunal-vaishnavi commented on 2023-11-01

sync and merge with latest main

5ce04d11

tianleiwu dismissed these changes on 2023-11-02

fix benchmark_all and change dist_settings mpi is only required for m…

dcddb283

frank-dong-ms dismissed their stale review via dcddb283 2 years ago

github-advanced-security commented on 2023-11-02

fix lint error

aa8ffb68

github-advanced-security commented on 2023-11-02

fix readme

4c567ca0

remove extra line

31c011a1

kunal-vaishnavi approved these changes on 2023-11-02

frank-dong-ms merged dabd395f into main 2 years ago

tianleiwu removed release:1.16.2

tianleiwu removed sdxl_llama

Reviewers

kunal-vaishnavi

tianleiwu

hanbitmyths

mindest

github-advanced-security

Assignees

No one assigned

Labels

None yet

Milestone

No milestone

onnxruntime llama 70b model fusion and shardding #18175 Merged

llama 70b model fusion and shardding #18175

onnxruntime
llama 70b model fusion and shardding
#18175

Merged