huggingface candle issues

huggingface / candle

Minimalist ML framework for Rust

Apache License 2.0

13.79k stars 750 forks source link

issues

Least commented

Newest Most commented Recently updated Oldest Least commented Least recently updated

Model Wishlist

#1177 LaurentMazare opened 6 months ago
58
Cannot run examples with --features cuda option

#353 dbrowne opened 9 months ago
48
Support for quantisation

#359 okpatil4u closed 2 months ago
36
Apple silicon (MPS backends) support?

#313 overheat closed 2 months ago
36
Llama models quantized with tensor-tools don't have llama.attention.head_count in metadata

#1182 soupslurpr closed 6 months ago
30
Reinforcement Learning

#1065 christopher-wilke closed 2 months ago
27
Quantized whisper example

#574 soupslurpr closed 7 months ago
26
I can't solve the porting error of ResNet

#1113 kingzcheung closed 6 months ago
25
Any tips to speed up quantized Whisper inference on Android?

#1048 soupslurpr opened 7 months ago
24
PyO3: Add `equal` and `__richcmp__` to `candle.Tensor`

#1099 LLukas22 closed 6 months ago
23
Quantized Implementations are slow

#1043 okpatil4u closed 2 months ago
23
Simple inference from loading traced pytorch models

#978 maxwellflitton opened 7 months ago
23
Quantized much slower than llama.cpp with same model and settings...

#1939 oddpxl opened 1 month ago
22
Curious obeservation with T5 example and Apple Accelerate

#868 okpatil4u closed 7 months ago
20
T5 inference in Candle

#621 shivance closed 8 months ago
20
AMD hardware support for training and Inference

#346 ksasi opened 9 months ago
20
Explicit panic on Falcon

#593 noahgift closed 8 months ago
19
[Examples] yi running with gpu enabled gives an error loading function: "Function 'cast_bf16_f32' does not exist"

#1532 fankaiLiu closed 4 months ago
17
Make the Python Wrapper more Hackable and simplify Quantization

#1010 LLukas22 closed 7 months ago
17
Make a tutorial for porting PyTorch models to candle

#1075 LaurentMazare closed 6 months ago
16
Metal Backend not properly loading large models at 16GB of RAM

#1568 bayedieng opened 4 months ago
15
Support Meta Nougat model

#1070 Christof23 opened 7 months ago
15
Recent revision for contiguous check has problems

#1991 guoqingbao opened 1 month ago
14
PReLU support?

#1342 joeyballentine closed 5 months ago
14
Performance issues compared to Pytorch

#1139 joeyballentine opened 6 months ago
14
Question: How to use quantized tensors?

#1006 EricLBuehler closed 7 months ago
14
How to fine-tune Llama?

#894 EricLBuehler closed 7 months ago
14
Math is difficult to implement

#691 timokoesters opened 8 months ago
14
Latest tensor squeeze impl make cuda matmal fail

#1948 yinqiwen closed 1 month ago
13
Poor generation when using quantised models on cuda

#1765 LaurentMazare closed 2 months ago
13
Llama is Incredibly slow at macbook air m2 16gb mem

#1757 charlesxsh opened 2 months ago
13
Metavoice - first cut

#1717 LaurentMazare closed 2 months ago
13
Slow generation compared to transformers + PyTorch

#1683 hugoabonizio opened 3 months ago
13
How to reduce memory usage of backpropagation?

#1241 laptou opened 6 months ago
13
`Module::set_training` takes hard-to-use `&mut self`

#783 EricLBuehler closed 8 months ago
13
Need train example

#637 evgenyigumnov closed 2 months ago
13
GPU not being used.

#540 SubinVidhu closed 8 months ago
13
Avoiding `.contiguous` call before `matmul`

#1965 EricLBuehler closed 1 month ago
12
Can't loop over model implementation based off examples more than N times (7-20+ it ends up breaking)

#1918 groovybits opened 1 month ago
12
Add Euler Ancestral Discrete Scheduler

#1390 edwin0cheng closed 5 months ago
12
Possible bug in `Tensor::matmul`

#811 EricLBuehler closed 8 months ago
12
Bigram Model

#406 okpatil4u closed 9 months ago
12
Quantized Matmul: Small batches are slower than no-batch

#2074 lucasavila00 closed 3 weeks ago
11
Interactive mode for Mistral example

#1896 evgenyigumnov closed 1 month ago
11
question: what GPU can run the mixtral example?

#1733 zwpaper opened 2 months ago
11
Remove unnecessary memset for reshape, concat and contiguous & support inplace op

#1680 guoqingbao closed 2 months ago
11
Quantized models on Cuda

#1250 EmilLindfors opened 6 months ago
11
Llama example uses ~10% of the CPU

#1103 kstavro opened 7 months ago
11
Benchmarking Speedups on M1 Max System

#756 okpatil4u closed 8 months ago
11
Architecture Support and Deployment Strategy

#642 noahgift closed 7 months ago
11