issues
search
FlagOpen
/
FlagGems
FlagGems is an operator library for large language models implemented in Triton Language.
Apache License 2.0
344
stars
47
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
bug fix of empty list of dimensions for sum
#328
FatJhon
opened
29 minutes ago
0
add program_id & num_programs that returns tensor in tl.int64
#327
iclementine
opened
1 hour ago
0
Fix masked fill
#326
GwokHiujin
closed
1 hour ago
0
Warning message needed in case rng is involved in gems
#325
tongxin
opened
2 hours ago
0
[benchmark] [no ci] update summary script
#324
StrongSpoon
closed
49 minutes ago
1
[Operator] group_norm and layer_norm don't support weight and bias as none
#323
StrongSpoon
opened
1 day ago
0
[Operator] Add sort op
#322
zhzhcookie
opened
1 day ago
0
Code Contribution: 【Lv2】【Operator Development】cummin
#321
StrongSpoon
opened
3 days ago
1
Code Contribution: 【Lv2】【Operator Development】cummax
#320
StrongSpoon
opened
3 days ago
1
Code Contribution: 【Lv2】【Operator Development】cumprod
#319
StrongSpoon
opened
3 days ago
1
Code Contribution: 【Lv2】【Operator Development】argmin
#318
StrongSpoon
opened
3 days ago
1
Code Contribution: 【Lv3】【Operator Development】gather_backward
#317
StrongSpoon
opened
3 days ago
0
Code Contribution: 【Lv3】【Operator Development】index_select_backward
#316
StrongSpoon
opened
3 days ago
0
Code Contribution: 【Lv3】【Operator Development】batch_norm_backward
#315
StrongSpoon
opened
3 days ago
1
Code Contribution: 【Lv1】【Operator Development】diagonal_backward
#314
StrongSpoon
opened
3 days ago
2
Code Contribution: 【Lv3】【Operator Development】istft
#313
StrongSpoon
opened
3 days ago
0
Code Contribution: 【Lv3】【Operator Development】 conv3d
#312
StrongSpoon
opened
3 days ago
1
[Fix] where errors on zero dim cpu tensor inputs
#311
tongxin
closed
4 days ago
0
Iluvatar
#310
awayzjj
closed
4 days ago
0
[Operator] Fix embedding op backward with bf16
#309
zhzhcookie
opened
5 days ago
0
fix max: numel exceeds triton maximum tensor numel #304
#308
haowen-han
closed
5 days ago
1
[Benchmark] change core shapes of diag to avoid oom on large shapes
#307
StrongSpoon
closed
5 days ago
0
[CI/CD] Move CI into a new machine into a docker
#306
Bowen12992
closed
4 days ago
0
在跑通『LLaVA单卡+gems』过程总遇到的embedding反向BF16的bug
#305
0x45f
opened
6 days ago
0
在跑通『LLaVA单卡+gems』过程总遇到的 max.dim超过tensor最大元素数的 bug
#304
0x45f
closed
5 days ago
2
在跑通『LLaVA单卡+gems』过程总遇到的where.self算子CPU bug
#303
0x45f
closed
4 days ago
0
在跑通『LLaVA单卡+gems』过程总遇到的cat除零bug
#302
0x45f
opened
6 days ago
1
在跑通『LLaVA单卡+gems』过程总遇到的3个框架集成bug
#301
0x45f
opened
1 week ago
4
maximum报错
#300
zhaogf01
opened
1 week ago
3
masked_fill op bug
#299
Salv1a
closed
1 week ago
2
Mthreads/master
#298
machuanjiang
opened
1 week ago
0
[bugFix] Support `masked_fill.Scalar`
#297
GwokHiujin
closed
1 week ago
0
[Operator] support instance_norm
#296
ZQPei
opened
1 week ago
0
[bugfix] mark eps as dns arg instead of constexpr
#295
awayzjj
closed
1 week ago
0
[muti_backend]
#294
Galaxy1458
opened
1 week ago
0
Obtain the maximum value of a two-dimensional discontinuous tensor without performing time-consuming continuous operations
#293
haowen-han
closed
4 days ago
5
[Operator] Add kron op
#292
growvv
opened
1 week ago
0
fix min and max, use loop_kernel when reduction size is larger than 16384
#291
iclementine
opened
1 week ago
0
blas tflops support
#290
shuailong616
opened
1 week ago
2
Narrow Benchmark Size for UnaryReductionBenchmark, Handle OOM Exception
#289
kiddyjinjin
closed
1 week ago
0
[Operator] Add diag_embed
#288
2niuhe
opened
1 week ago
3
Add quantile op
#287
CelysPr
opened
2 weeks ago
2
Update README.md
#286
Tango2018cc
closed
6 days ago
2
About performance between ATen and triton
#285
gavin838
closed
2 weeks ago
4
Add filter_shape method to Benchmark class
#284
Gxiandy
closed
1 week ago
3
[Operator] Add diag op and related tests(#252)
#283
Gxiandy
closed
1 week ago
0
about "transpose" and "permute"
#282
Shaquille-Wu
closed
1 week ago
2
Non-code Contribution: Triton Blog 3
#279
Tango2018cc
opened
2 weeks ago
1
Non-code Contribution: Triton Blog 2
#278
Tango2018cc
opened
2 weeks ago
1
Non-code Contribution: Triton Blog 1
#277
Tango2018cc
opened
2 weeks ago
1
Next