ggerganov ggml Q A · Discussions · GitHub

Sort by: Latest activity

Categories

Community guidelines

Q&A Discussions

Ask the community for help

You must be logged in to vote

[Low Priority] Wonder why tensor scheme in gguf is an enum rather than a randomly selected 64bit (or a UUID)

mofosyne asked May 20, 2024 in Q&A · Unanswered

2
You must be logged in to vote

GGML vs XNNPACK

Techget asked May 15, 2024 in Q&A · Unanswered

0
You must be logged in to vote

Intemittent memory errors when trying to load model.

balisujohn asked May 7, 2024 in Q&A · Answered

3
You must be logged in to vote

Is there an efficient way to do tensor-scalar addition?

balisujohn asked May 3, 2024 in Q&A · Answered

1
You must be logged in to vote

Using a KV cache with GGML

benraha asked Apr 10, 2024 in Q&A · Unanswered

1
You must be logged in to vote

Encodec-cpp-ggml: when I make the Encodec (cpp) version, I am using ggml ggml-model-f16.bin, GGML_TYPE_COUNT miss match

monircsueb asked Apr 5, 2024 in Q&A · Closed · Unanswered

0
You must be logged in to vote

How can I do equivalent of torch.chunk in ggml tensor ?

mnlife asked Apr 1, 2024 in Q&A · Unanswered

1
You must be logged in to vote

How bcast works

hipudding asked Mar 21, 2024 in Q&A · Closed · Unanswered

0
You must be logged in to vote

Should there be a matrix check in ggml_transpose?

crnsh asked Mar 3, 2024 in Q&A · Answered

1
You must be logged in to vote

Is there a slice operator in ggml? If not, how do I use it / implement it?

crnsh asked Mar 2, 2024 in Q&A · Answered

1
You must be logged in to vote

Conversion older ggml code to latest version

HoiV asked Feb 19, 2024 in Q&A · Answered

1
You must be logged in to vote

Adaptive lora - some questions on ggml

cmp-nct asked Feb 15, 2024 in Q&A · Unanswered

0
You must be logged in to vote

Is there a recommended way to implement an analogous operation to torch.where?

balisujohn asked Feb 1, 2024 in Q&A · Closed · Unanswered

1
You must be logged in to vote

minibatch size and ADAM optimizer

bssrdf asked Jan 11, 2024 in Q&A · Closed · Unanswered

0
You must be logged in to vote

How to optimize more than one loss functions?

bssrdf asked Jan 14, 2024 in Q&A · Closed · Unanswered

0
You must be logged in to vote

How to transpose a tensor and have the transpose reflected in contiguous memory with CUDA backend

balisujohn asked Jan 7, 2024 in Q&A · Answered

3
You must be logged in to vote

Is there a way to downcast a float32 tensor to a float16 tensor?

balisujohn asked Jan 1, 2024 in Q&A · Answered

1
You must be logged in to vote

Need help to understand the (backward) computational graph

bssrdf asked Dec 22, 2023 in Q&A · Answered

2
You must be logged in to vote

[REQUEST] Tips for implementing custom models with ggml

astariul asked Dec 15, 2023 in Q&A · Answered

3
You must be logged in to vote

How FP16 embeddings are converted to FP32 tensors in GPT-2 example ?

astariul asked Dec 14, 2023 in Q&A · Answered

1
You must be logged in to vote

Llama.cpp a lot slower when using cmake compared to using w64devkit

v4lentin1879 asked Oct 26, 2023 in Q&A · Unanswered

3
You must be logged in to vote

running inference in parallel in multiple threads

nerdypepper asked Oct 10, 2023 in Q&A · Closed · Answered

1
You must be logged in to vote

Why memory usage not change when add different input with GGML format

SiraHaruethaipree asked Oct 10, 2023 in Q&A · Unanswered

0
You must be logged in to vote

why?

alph4b3th asked Sep 28, 2023 in Q&A · Unanswered

0
You must be logged in to vote

Why GGUF supports mmap specially?

AsakusaRinne asked Aug 28, 2023 in Q&A · Unanswered

3