Giters
ggerganov
/
ggml
Tensor library for machine learning
Geek Repo:
Geek Repo
Github PK Tool:
Github PK Tool
Stargazers:
9876
Watchers:
120
Issues:
364
Forks:
899
ggerganov/ggml Issues
Zig build is broken
Updated
6 months ago
Comments count
2
gguf : opening an invalid file may cause a out of bounds access
Closed
6 months ago
[Question] How do I force a computation of a tensor/force a dependency between 2 tensors?
Closed
6 months ago
Comments count
2
the gguf module in ggml.c must have an option to write as a stream.
Updated
6 months ago
Compilation error with make on Linux Lite related to AVX.
Closed
6 months ago
Comments count
1
...
Closed
6 months ago
Question about convert auto-gptq model to ggml format?
Updated
6 months ago
[2GPU] Memcpy2D of matrixXmatrix -- src size (and form)
Updated
7 months ago
Comments count
3
test-conv-transpose fails when building with sanitizers enabled
Closed
7 months ago
Issue with ggml
Updated
7 months ago
Comments count
1
Using memcpy on tensor data when the tensor is not contiguous
Updated
7 months ago
Comments count
1
How is the data allocated by ggml_allocr_alloc align?
Closed
7 months ago
Comments count
2
ggml : improve memory allocation for weights and similar lists of tensors
Closed
4 months ago
Comments count
2
Build fails on Centos7 with make: *** [all] Error 2
Closed
7 months ago
Comments count
5
ggml : create a complete document
Updated
6 months ago
Comments count
5
ggml : remove GGML_MAX_NODES limit
Closed
7 months ago
Comments count
10
Question about ggml-alloc assert in CPU ggml-backend version of Sam.cpp
Closed
7 months ago
Comments count
6
ggml : replace conv stage_0 and stage_1 with im2col and mul_mat
Closed
6 months ago
Comments count
10
Conv2D kernel CuBLAS implementation - need feedback
Closed
6 months ago
Comments count
11
ggml_allocr_new null pointer exception
Closed
7 months ago
ggml : expose hash table API from ggml.c and reuse in ggml-alloc
Closed
7 months ago
Comments count
17
CUDA implementation of ggml_clamp
Closed
8 months ago
Comments count
3
[Question] What is the status of Vulkan backend?
Updated
2 months ago
Comments count
12
How about adding "tokenizer.ggml.cls_token_id" for special cls token.
Updated
8 months ago
Issue inferencing HuggingFace's GPT-J 4 bits model
Updated
7 months ago
Comments count
1
[Feature Request] any hyper-resolution inference
Updated
7 months ago
Comments count
2
Is T5 (mLongT5, FlanT5 etc.) being developed with GGML ?
Updated
7 months ago
Comments count
1
prompt is too long (539 tokens, max 508)
Updated
8 months ago
Can't some operations run in parallel?
Updated
7 months ago
Comments count
1
Merge HF LoRa adapter with a quantized GPT-J model using ggml
Updated
8 months ago
Port k-quants support from ggerganov/llama.cpp to ggerganov/ggml
Updated
8 months ago
Asserting over nb[0] as type check causes issue for tensors after permuting
Updated
8 months ago
Comments count
1
Quantized matmul with CUDA sets the result to zero instead of properly computing it
Closed
8 months ago
Comments count
5
Why GGML_F16_STEP is 32?
Updated
7 months ago
Comments count
3
Why GPT-J performs better on graviton without using simd than x86 using simd
Updated
8 months ago
Converting an arbritrary HF Transformer GPT2 to ggml format
Updated
8 months ago
Comments count
1
No output from executable when building with MSYS2
Updated
8 months ago
Error when using ggml_scale
Closed
8 months ago
Comments count
4
Directly converted from bfloat16 weights are 20x slower than converted from float32 ones.
Updated
8 months ago
sam visual studio
Closed
8 months ago
How to access the ggml_tensor data?
Closed
8 months ago
Comments count
1
sam bad result
Closed
8 months ago
Comments count
8
How to easily convert Meta's MMS-ASR models to ggml?
Updated
8 months ago
Comments count
1
ggml : better way to express implicit node dependencies in a graph
Updated
8 months ago
linesearch_backtracking has confusing logic
Closed
8 months ago
Comments count
1
Confirmation about the order of tensor dimensions
Closed
8 months ago
Comments count
5
mpt-1b fails with mpt_model_load: unknown tensor 'transformer.blocks.0.attn.k_ln.weight' in model file
Updated
8 months ago
Comments count
4
[Feature request] Add support/demo implementation for Qwen-VL GGUF model
Updated
7 months ago
Comments count
2
Use custom GPT-J checkpoint
Updated
8 months ago
Comments count
1
gpt-j, starcoder, gptneox examples cause "not enough space in the context's memory pool" for batches >32
Updated
9 months ago
Comments count
2
Previous
Next