vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

ggml : add IQ2 to test-backend-ops + refactoring (#4990)

Browse source 
* ggml : add IQ2 to test-backend-ops + refactoring

ggml-ci

* cuda : update supports_op for IQ2

ggml-ci

* ci : enable LLAMA_CUBLAS=1 for CUDA nodes

ggml-ci

* cuda : fix out-of-bounds-access in `mul_mat_vec_q`

ggml-ci

* tests : avoid creating RNGs for each Q tensor

ggml-ci

* tests : avoid creating RNGs for each tensor

ggml-ci
This commit is contained in:
Georgi Gerganov 2024-01-17 18:54:56 +02:00 • committed by GitHub
parent ba69bbc84c
commit 38566680cd
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
9 changed files with 128 additions and 87 deletions
Download patch file Download diff file Expand all files Collapse all files

3
ggml-quants.h
View file

@ -257,3 +257,6 @@ size_t quantize_q4_0 (const float * src, void * dst, int nrows, int n_per_row,
size_t quantize_q4_1 (const float * src, void * dst, int nrows, int n_per_row, int64_t * hist, const float * imatrix);
size_t quantize_q5_0 (const float * src, void * dst, int nrows, int n_per_row, int64_t * hist, const float * imatrix);
size_t quantize_q5_1 (const float * src, void * dst, int nrows, int n_per_row, int64_t * hist, const float * imatrix);
void iq2xs_init_impl(int grid_size);
void iq2xs_free_impl(int grid_size);