issues
search
intel
/
xFasterTransformer
Apache License 2.0
322
stars
56
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
[Kernel] Add GPU kernels and enable LLaMA model.
#372
changqi1
closed
1 month 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
Fix Qwen prompt.json
#368
JunxiChhen
closed
2 months ago
2
[Common] Modify resize() in DecoderContext to support
#367
pujiang2018
closed
2 months ago
0
[Model] add interface for seq meta.
#366
Duyi-Wang
closed
2 months ago
0
[Layers] fix build error
#365
abenmao
closed
2 months ago
0
[Models] Add AttnMetaData and fix attn build error
#364
abenmao
closed
2 months ago
0
[Common] Refactor sequence.h.
#363
Duyi-Wang
closed
2 months ago
0
[Util] Remove DecoderContext in computeSoftmax
#362
pujiang2018
closed
2 months ago
0
[Kernels] Refactor flash attention for continuous batching.
#361
abenmao
closed
2 months ago
0
[Benchmark] Calculate throughput using avg latency.
#360
Duyi-Wang
closed
2 months ago
0
[GPU] Add GPU build option.
#359
changqi1
closed
2 months ago
0
[Model] Fix compile error of embeddingForward in YaRNLlama
#358
pujiang2018
closed
2 months ago
0
[Framework] Continuous Batching Support
#357
pujiang2018
closed
2 months ago
0
[Common] Add sampling params into group seq.
#356
Duyi-Wang
closed
2 months ago
1
[Model] Achieve whole pipeline parallel.
#355
changqi1
opened
2 months ago
1
[Fix] add utf-8 encoding.
#354
marvin-Yu
closed
2 months ago
0
[Layer] Remove unused functions in Decoder layer
#353
pujiang2018
closed
2 months ago
0
[Version] v1.6.0.
#352
Duyi-Wang
closed
2 months ago
0
[Common] Move Matrix into xft namespace.
#351
Duyi-Wang
closed
2 months ago
0
[Layer][Kernel] Merge batchSize and seqLen into one param (tokenSize) in TokenEembedding
#350
pujiang2018
closed
2 months ago
0
[UT] Remove beam search test temporarily.
#349
Duyi-Wang
closed
2 months ago
0
[Kernel][UT] Kernel impl. of crossAttnByHead and unit test for cross attention.
#348
pujiang2018
closed
2 months ago
0
[Evaluation] fix the model register bug in evaluation
#347
abenmao
closed
2 months ago
0
[Kernel] Add 'acc' param in small_gemm, add lacked and remove unused small_gemm kernels.
#346
pujiang2018
closed
3 months ago
0
The current commit(d666741) of xFT make failed.
#345
xiuying1
closed
2 months ago
3
[Models] YaRN-Llama full-link bf16 support
#344
abenmao
closed
2 months ago
0
[Common] Add sequenceMeta, sequenceGroup and sequenecePool.
#343
changqi1
closed
3 months ago
0
[xDNN] Release v1.4.6.
#342
changqi1
closed
3 months ago
0
Build Error: Failure to Download and Configure xdnn_lib
#341
Damonpkl
closed
3 months ago
1
[model] Add llama3 model.
#340
marvin-Yu
closed
3 months ago
0
performance issue for opt-1.3b with BS=1 BF16
#339
bin1guo
opened
3 months ago
1
[Demo] Add kvcache type option in web demo.
#338
Duyi-Wang
closed
3 months ago
0
[Benchmark] Add KVCache data type option.
#337
Duyi-Wang
closed
3 months ago
1
[KVCache] Add inferface and register for kvcache.
#336
Duyi-Wang
closed
3 months ago
0
chatglm3 6b error
#335
zhm-algo
opened
3 months ago
1
[UT] Add unit test for xft::crossAttnShardedHead
#334
pujiang2018
closed
3 months ago
0
[Sampling] Decouple greedy search from searcher.
#333
Duyi-Wang
closed
2 months ago
0
[Layer] Add SequenceMeta, SequencePool and init pipeline parrallel function.
#332
changqi1
closed
3 months ago
1
[RAEDME] Update readme for the dependent lib.
#331
xwang98
closed
3 months ago
0
[Model] Add Qwen2 model.
#330
marvin-Yu
closed
3 months ago
2
[KVCache] KVCache and KVCacheMgr refactor to support continuous batching.
#329
pujiang2018
closed
2 months ago
1
[bug] Met some problems while following *step by step tutorial*
#328
lum1n0us
closed
3 months ago
2
[Finetune] Scripts for Llama2-7b lora finetune example using stock pytorch
#327
ustcuna
closed
2 months ago
2
[Sample] Fix numeric overflow when calculate softmax.
#326
Duyi-Wang
closed
3 months ago
1
[Eval] Add eval test with opencompass.
#325
marvin-Yu
opened
3 months ago
1
[Doc] Add develop docs.
#324
marvin-Yu
closed
3 months ago
0
[Layers] fix assert bug when concat gate&up
#323
abenmao
closed
3 months ago
0
Previous
Next