issues
search
BlackSamorez
/
tensor_parallel
Automatically split your PyTorch models on multiple GPUs for training & inference
MIT License
629
stars
39
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Does tensor_parallel support the model inference concurrently or in multi-threads?
#86
zoubaihan
closed
1 year ago
2
Does tensor_parallel support data parallel and tensor parallel hybrid training?
#85
liguodongiot
opened
1 year ago
0
Does tensor_parallel support multi-node tensor parallel training?
#84
liguodongiot
opened
1 year ago
6
Can I parallelize just one large layer?
#83
chinmayjog13
opened
1 year ago
1
Actually using SplitInsideChunks for gpt2
#82
BlackSamorez
closed
1 year ago
0
Request to fix the content about parallelformers in README.
#81
hyunwoongko
closed
1 year ago
1
Support for PEFT LoRA and 4-bit quantization
#80
morecry
closed
1 year ago
6
Not work with 4bit quant
#79
laoda513
closed
1 year ago
6
tp.convert_state_dict readme example fix
#78
BlackSamorez
closed
1 year ago
0
Error in README.Md, hence not able to load model with limited memory.
#77
vishakudupa
closed
1 year ago
5
Torch version requirement
#76
treya-lin
closed
1 year ago
4
Great work！ and can this work with deepspeedzero?
#75
laoda513
opened
1 year ago
0
Huggingface Accelerate
#74
conceptofmind
closed
1 year ago
1
State dict fixes for tied weights
#73
BlackSamorez
closed
1 year ago
1
What is the difference between this project and autotp of deepspeed?
#72
frankxyy
closed
1 year ago
1
cuda memory not evenly distributed between devices
#71
frankxyy
closed
1 year ago
6
Torch distributed hotfix
#70
BlackSamorez
closed
1 year ago
1
set distributed=True, return AttributeError: 'NoneType' object
#69
hijeffwu
closed
1 year ago
2
Peft LoRA support
#68
BlackSamorez
closed
1 year ago
1
How to load lora weights？
#67
Vincent131499
closed
1 year ago
13
Slow inference performance for large Llama models compared to naive MP
#66
sgsdxzy
opened
1 year ago
26
Mention linear speedup in Readme
#65
BlackSamorez
closed
1 year ago
0
Set seed for tests reproducibility
#64
BlackSamorez
closed
1 year ago
1
Small readme patch
#63
BlackSamorez
closed
1 year ago
0
Added int8 LLMs demo link
#62
BlackSamorez
closed
1 year ago
0
CodeGen config
#61
BlackSamorez
closed
1 year ago
0
Converting state dicts without model creation
#60
BlackSamorez
closed
1 year ago
3
New version for dispatch hotfix
#59
BlackSamorez
closed
1 year ago
0
Shard parameters initial dispatch fix
#58
BlackSamorez
closed
1 year ago
0
Unpersistent buffers meta loading fix
#57
BlackSamorez
closed
1 year ago
1
_reorder_cache fix for generation utils
#56
BlackSamorez
closed
1 year ago
0
GPT NeoX config
#55
BlackSamorez
closed
1 year ago
0
Meta devices support
#54
BlackSamorez
closed
1 year ago
1
LLaMa models
#53
BlackSamorez
closed
1 year ago
0
Removing accelerate hooks before splitting the model
#52
BlackSamorez
closed
1 year ago
0
Support LLaMA Models, including HuggingFace-adapted variants
#51
dustydecapod
closed
1 year ago
7
Version update
#50
BlackSamorez
closed
1 year ago
0
Saving utilities
#49
BlackSamorez
closed
1 year ago
1
How to use trained models?
#48
Den4ikAI
closed
1 year ago
3
Adding support for more model architectures
#47
BlackSamorez
closed
1 year ago
2
False negative test results for test_convs. Flaky test
#46
BlackSamorez
closed
1 year ago
1
Add more predefined configs
#45
BlackSamorez
opened
1 year ago
0
Replace architecture with model_type
#44
BlackSamorez
closed
1 year ago
0
Config refactoring
#43
BlackSamorez
closed
1 year ago
4
New ideas
#42
aizamaksutova
opened
1 year ago
0
GPU Contention
#41
aizamaksutova
opened
1 year ago
0
Fixed PyPi link in readme
#40
BlackSamorez
closed
1 year ago
0
Version update
#39
BlackSamorez
closed
1 year ago
0
_TensorParallelWrapper attribute forwarding
#38
BlackSamorez
closed
1 year ago
0
hotfix canonic torch.device
#37
justheuristic
closed
1 year ago
0
Previous
Next