issues
search
intel
/
xFasterTransformer
Apache License 2.0
270
stars
53
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Add --padding and fix bug
#418
yangkunx
closed
1 month ago
0
[Kernel] Add oneDNN AMX_FP16 compute kernels.
#417
changqi1
closed
1 month ago
1
[Dependency] Update torch to 2.3.0.
#416
Duyi-Wang
closed
1 month ago
0
[Kernel] Add FP16 MHA and MLP kernels.
#415
changqi1
closed
1 month ago
0
Can we summarize the meanings of data type like bf16_fp16?, for example, what's activation data type and output data type, what's the computing instruction?
#414
heagoo
opened
1 month ago
1
[output issue] found mistakes in llama-3-70b output by bf16_int4 during benchmark
#413
intelyoungway
opened
1 month ago
1
[Kenrel] Add FP16 LLaMA YARN rotary_embedding.
#412
changqi1
closed
1 month ago
1
qwen1.5-32b long text input issue
#411
zhm-algo
closed
2 weeks ago
5
[xDNN] Release v1.5.0.
#410
changqi1
closed
1 month ago
0
[Benchmark] Add platform options. Support real model.
#409
JunxiChhen
closed
1 month ago
1
[Kernel] Add FP16 rmsnorm and rope kernels.
#408
changqi1
closed
1 month ago
0
[Models/Layers/Kernels] Add Baichuan1/2 full-link bf16 support & Fix next-tok gen bug
#407
abenmao
closed
1 month ago
0
[Example] Fix incorrect tensor dimension with latest interface
#406
pujiang2018
closed
1 month ago
0
[Bug] fix incorrect input offset computing
#405
pujiang2018
closed
1 month ago
0
[Interface] Support List[int] and List[List[int]] for set_input_sb.
#404
Duyi-Wang
closed
1 month ago
0
llama-2-70b memory usage
#403
zhm-algo
closed
1 month ago
2
[Layers] Add alibiSlopes Attn && Flash Attn for CB.
#402
abenmao
closed
1 month ago
0
[Example] Add demo of offline continuous batching
#401
pujiang2018
closed
1 month ago
0
[Interface] Change return shape of forward_cb.
#400
Duyi-Wang
closed
1 month ago
0
[KVCache] Remove FP32 data type.
#399
Duyi-Wang
closed
1 month ago
0
[Interface] Add python api for continuous batching.
#398
Duyi-Wang
closed
1 month ago
0
[Layer] Better method to reinterpret KV cache
#397
pujiang2018
closed
1 month ago
0
[API] Optimize API Impl.
#396
changqi1
closed
1 month ago
0
What's the meaning of bf16_int4 datatype?
#395
LeiZhou-97
closed
1 month ago
4
[Model] Check maxLen should be [input len, model max len].
#394
Duyi-Wang
closed
1 month ago
0
[Example] More check in C++ continuous batching example
#393
pujiang2018
closed
1 month ago
0
[Example] Fix continuous batching C++ example.
#392
Duyi-Wang
closed
1 month ago
0
[Bug] Fix incorrect buffer size calculation
#391
pujiang2018
closed
1 month ago
0
[Example] add cb_check example
#390
pujiang2018
closed
1 month ago
0
[Model][Layer] Correct output of the new forward
#389
pujiang2018
closed
1 month ago
0
[Build] Fix namespace build issue.
#388
Duyi-Wang
closed
1 month ago
0
[Common] DecoderContext::resize bug fix
#387
pujiang2018
closed
1 month ago
0
[API] Add LLaMA decoder API.
#386
changqi1
closed
1 month ago
3
[Demo] Add abbreviation for output length.
#385
Duyi-Wang
closed
1 month ago
0
[Layer] update mlp for CB.
#384
marvin-Yu
closed
1 month ago
0
[Layers] Added RotaryEmbedding forward for bc mode & Fixed rope ut
#383
abenmao
closed
1 month ago
0
[Layer] Cross attention impl. for CB
#382
pujiang2018
closed
1 month ago
0
[Framework] Update set_input for cb.
#381
Duyi-Wang
closed
1 month ago
0
llama-2-7B benchmarking error with chinese prompts
#380
qdym188
closed
1 month ago
1
[Framework] Code fix to make new path for CB work
#379
pujiang2018
closed
2 months ago
0
[API] Add LLaMA attention API.
#378
changqi1
closed
1 month ago
1
[Model] Return seqIDs when set input.
#377
Duyi-Wang
closed
2 months ago
0
[Sampling] Add greedy search for cb path.
#376
Duyi-Wang
closed
2 months ago
0
[Model/Layer] New forward to support CB (CommonDecoder->DecoderBlock->DecoderLayer->Attention/MLP)
#375
pujiang2018
closed
2 months ago
1
[CMake] Remove evaluation under XFT_BUILD_TESTS option.
#374
Duyi-Wang
closed
2 months ago
0
[Sampling] Add repetition penalty for new seq type.
#373
Duyi-Wang
closed
2 months ago
0
[Kernel] Add GPU kernels and enable LLaMA model.
#372
changqi1
closed
3 weeks ago
0
[Common] New KVCacheMgr to support CB
#371
pujiang2018
closed
2 months ago
1
[Model] Fix ICX build issue.
#370
changqi1
closed
2 months ago
0
[Model] New CommonDecoder::forward impl. skeleton
#369
pujiang2018
closed
2 months ago
2
Previous
Next