Skip to content

IQ5_KS_R4: row-interleaved IQ5_KS #426

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 5 commits into from
May 17, 2025
Merged

IQ5_KS_R4: row-interleaved IQ5_KS #426

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
48ebde8
iq5_ks_r4: basics
May 15, 2025
25ce0f0
iq5_ks_r4: Zen4 works
May 15, 2025
2b040ae
iq5_ks_r4: AVX2 works
May 15, 2025
2881f5f
iq5_ks_r4: NEON
May 16, 2025
9dd452d
Fix iq5_ks on NEON
May 16, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions examples/quantize/quantize.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -67,6 +67,7 @@ static const std::vector<struct quant_option> QUANT_OPTIONS = {
{ "IQ4_XS", LLAMA_FTYPE_MOSTLY_IQ4_XS, " 4.25 bpw non-linear quantization", },
{ "IQ4_KS", LLAMA_FTYPE_MOSTLY_IQ4_KS, " 4.25 bpw non-linear quantization", },
{ "IQ4_KS_R4",LLAMA_FTYPE_MOSTLY_IQ4_KS_R4,"IQ4_KS repacked", },
{ "IQ5_KS_R4",LLAMA_FTYPE_MOSTLY_IQ5_KS_R4,"IQ5_KS repacked", },
{ "IQ4_KSS", LLAMA_FTYPE_MOSTLY_IQ4_KSS, " 4.0 bpw non-linear quantization", },
{ "IQ5_KS", LLAMA_FTYPE_MOSTLY_IQ5_KS, " 5.25 bpw non-linear quantization", },
{ "IQ2_K", LLAMA_FTYPE_MOSTLY_IQ2_K, " 2.375 bpw non-linear quantization",},
Expand Down
2 changes: 2 additions & 0 deletions ggml/include/ggml.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -452,6 +452,7 @@ extern "C" {
GGML_TYPE_IQ4_K_R4 = 339,
GGML_TYPE_IQ5_K_R4 = 340,
GGML_TYPE_IQ4_KS_R4 = 344,
GGML_TYPE_IQ5_KS_R4 = 352,
GGML_TYPE_Q8_KV_R8 = 398,
GGML_TYPE_Q8_K_R8 = 399,
GGML_TYPE_COUNT,
Expand Down Expand Up @@ -540,6 +541,7 @@ extern "C" {
GGML_FTYPE_MOSTLY_IQ4_K_R4 = 332, // except 1d tensors
GGML_FTYPE_MOSTLY_IQ5_K_R4 = 333, // except 1d tensors
GGML_FTYPE_MOSTLY_IQ4_KS_R4 = 337, // except 1d tensors
GGML_FTYPE_MOSTLY_IQ5_KS_R4 = 341, // except 1d tensors
GGML_FTYPE_MOSTLY_Q8_KV_R8 = 398, // except 1d tensors
GGML_FTYPE_MOSTLY_Q8_K_R8 = 399, // except 1d tensors
};
Expand Down
7 changes: 7 additions & 0 deletions ggml/src/ggml-common.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -694,6 +694,13 @@ typedef struct {
} block_iq5_ks;
static_assert(sizeof(block_iq5_ks) == QK_K/32 + QK_K/2 + QK_K/8, "wrong iq5_ks block size/padding");

typedef struct {
uint8_t scales[QK_K/8];
uint8_t qs[QK_K*2];
uint8_t qh[QK_K/2];
} block_iq5_ks_r4;
static_assert(sizeof(block_iq5_ks_r4) == 4*sizeof(block_iq5_ks), "wrong iq5_ks_r4 block size/padding");


#endif // GGML_COMMON_DECL
#endif // GGML_COMMON_DECL
Expand Down
1 change: 1 addition & 0 deletions ggml/src/ggml-quants.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15451,6 +15451,7 @@ bool ggml_validate_row_data(enum ggml_type type, const void * data, size_t nbyte
case GGML_TYPE_IQ4_K_R4: break;
case GGML_TYPE_IQ5_K_R4: break;
case GGML_TYPE_IQ4_KS_R4:break;
case GGML_TYPE_IQ5_KS_R4:break;
case GGML_TYPE_Q8_KV_R8: break;
case GGML_TYPE_Q8_K_R8: break;
case GGML_TYPE_Q8_KV: break;
Expand Down
26 changes: 26 additions & 0 deletions ggml/src/ggml.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1339,6 +1339,23 @@ static const ggml_type_traits_t type_traits[GGML_TYPE_COUNT] = {
.vec_dot_type = GGML_TYPE_Q8_K32,
#else
.vec_dot_type = GGML_TYPE_Q8_K,
#endif
.nrows = 1,
.row_meta_size = 4,
},
[GGML_TYPE_IQ5_KS_R4] = {
.type_name = "iq5_ks_r4",
.blck_size = QK_K,
.type_size = sizeof(block_iq5_ks),
.is_quantized = true,
.to_float = (ggml_to_float_t) dequantize_row_iq5_ks_r4,
.from_float = quantize_row_iq5_ks_r4,
.from_float_ref = (ggml_from_float_t)quantize_row_iq5_ks_r4_ref,
.vec_dot = vec_dot_iq5_ks_r4_q8_k,
#if defined __AVX2__
.vec_dot_type = GGML_TYPE_Q8_K32,
#else
.vec_dot_type = GGML_TYPE_Q8_K,
#endif
.nrows = 1,
.row_meta_size = 4,
Expand Down Expand Up @@ -4478,6 +4495,7 @@ enum ggml_type ggml_ftype_to_ggml_type(enum ggml_ftype ftype) {
case GGML_FTYPE_MOSTLY_IQ4_XS: wtype = GGML_TYPE_IQ4_XS; break;
case GGML_FTYPE_MOSTLY_IQ4_KS: wtype = GGML_TYPE_IQ4_KS; break;
case GGML_FTYPE_MOSTLY_IQ4_KS_R4: wtype = GGML_TYPE_IQ4_KS_R4;break;
case GGML_FTYPE_MOSTLY_IQ5_KS_R4: wtype = GGML_TYPE_IQ5_KS_R4;break;
case GGML_FTYPE_MOSTLY_IQ4_KSS: wtype = GGML_TYPE_IQ4_KSS; break;
case GGML_FTYPE_MOSTLY_IQ5_KS: wtype = GGML_TYPE_IQ5_KS; break;
case GGML_FTYPE_MOSTLY_IQ2_K: wtype = GGML_TYPE_IQ2_K; break;
Expand Down Expand Up @@ -11242,6 +11260,7 @@ static void ggml_compute_forward_add(
case GGML_TYPE_IQ4_XS:
case GGML_TYPE_IQ4_KS:
case GGML_TYPE_IQ4_KS_R4:
case GGML_TYPE_IQ5_KS_R4:
case GGML_TYPE_IQ4_KSS:
case GGML_TYPE_IQ5_KS:
case GGML_TYPE_IQ2_K:
Expand Down Expand Up @@ -11715,6 +11734,7 @@ static void ggml_compute_forward_add1(
case GGML_TYPE_IQ4_XS:
case GGML_TYPE_IQ4_KS:
case GGML_TYPE_IQ4_KS_R4:
case GGML_TYPE_IQ5_KS_R4:
case GGML_TYPE_IQ4_KSS:
case GGML_TYPE_IQ5_KS:
case GGML_TYPE_IQ2_K:
Expand Down Expand Up @@ -11885,6 +11905,7 @@ static void ggml_compute_forward_acc(
case GGML_TYPE_IQ4_XS:
case GGML_TYPE_IQ4_KS:
case GGML_TYPE_IQ4_KS_R4:
case GGML_TYPE_IQ5_KS_R4:
case GGML_TYPE_IQ4_KSS:
case GGML_TYPE_IQ5_KS:
case GGML_TYPE_IQ2_K:
Expand Down Expand Up @@ -15382,6 +15403,7 @@ static void ggml_compute_forward_out_prod(
case GGML_TYPE_IQ4_XS:
case GGML_TYPE_IQ4_KS:
case GGML_TYPE_IQ4_KS_R4:
case GGML_TYPE_IQ5_KS_R4:
case GGML_TYPE_IQ4_KSS:
case GGML_TYPE_IQ5_KS:
case GGML_TYPE_IQ2_K:
Expand Down Expand Up @@ -15792,6 +15814,7 @@ static void ggml_compute_forward_set(
case GGML_TYPE_IQ4_XS:
case GGML_TYPE_IQ4_KS:
case GGML_TYPE_IQ4_KS_R4:
case GGML_TYPE_IQ5_KS_R4:
case GGML_TYPE_IQ4_KSS:
case GGML_TYPE_IQ5_KS:
case GGML_TYPE_IQ2_K:
Expand Down Expand Up @@ -16108,6 +16131,7 @@ static void ggml_compute_forward_get_rows(
case GGML_TYPE_IQ4_XS:
case GGML_TYPE_IQ4_KS:
case GGML_TYPE_IQ4_KS_R4:
case GGML_TYPE_IQ5_KS_R4:
case GGML_TYPE_IQ4_KSS:
case GGML_TYPE_IQ5_KS:
case GGML_TYPE_IQ2_K:
Expand Down Expand Up @@ -16741,6 +16765,7 @@ static void ggml_compute_forward_clamp(
case GGML_TYPE_IQ4_XS:
case GGML_TYPE_IQ4_KS:
case GGML_TYPE_IQ4_KS_R4:
case GGML_TYPE_IQ5_KS_R4:
case GGML_TYPE_IQ4_KSS:
case GGML_TYPE_IQ5_KS:
case GGML_TYPE_IQ2_K:
Expand Down Expand Up @@ -23810,6 +23835,7 @@ size_t ggml_quantize_chunk(
case GGML_TYPE_IQ4_XS: result = quantize_iq4_xs (src + start, (char *) dst + start_row * row_size, nrows, n_per_row, imatrix); break;
case GGML_TYPE_IQ4_KS: result = quantize_iq4_ks (src + start, (char *) dst + start_row * row_size, nrows, n_per_row, imatrix); break;
case GGML_TYPE_IQ4_KS_R4:result = quantize_iq4_ks_r4(src + start, (char *) dst + start_row * row_size, nrows, n_per_row, imatrix); break;
case GGML_TYPE_IQ5_KS_R4:result = quantize_iq5_ks_r4(src + start, (char *) dst + start_row * row_size, nrows, n_per_row, imatrix); break;
case GGML_TYPE_IQ4_KSS: result = quantize_iq4_kss(src + start, (char *) dst + start_row * row_size, nrows, n_per_row, imatrix); break;
case GGML_TYPE_IQ5_KS: result = quantize_iq5_ks (src + start, (char *) dst + start_row * row_size, nrows, n_per_row, imatrix); break;
case GGML_TYPE_IQ2_K: result = quantize_iq2_k (src + start, (char *) dst + start_row * row_size, nrows, n_per_row, imatrix); break;
Expand Down
Loading