nikola/llama_cpp
1
0
Fork 0
Code Issues Pull Requests Actions 33 Packages Projects Releases Wiki Activity

model: add support for qwen3vl series (#16780)

Browse Source 
* support qwen3vl series.

Co-authored-by: Thireus ☠ <Thireus@users.noreply.github.com>
Co-authored-by: yairpatch <yairpatch@users.noreply.github.com>
Co-authored-by: LETS-BEE <LETS-BEE@users.noreply.github.com>

* bugfix: fix the arch check for qwen3vl-moe.

* use build_ffn

* optimize deepstack structure

* optimize deepstack feature saving

* Revert "optimize deepstack feature saving" for temporal fix

This reverts commit f321b9fdf13e59527408152e73b1071e19a87e71.

* code clean

* use fused qkv in clip

* clean up / rm is_deepstack_layers for simplification

* add test model

* move test model to "big" section

* fix imrope check

* remove trailing whitespace

* fix rope fail

* metal : add imrope support

* add imrope support for sycl

* vulkan: add imrope w/o check

* fix vulkan

* webgpu: add imrope w/o check

* Update gguf-py/gguf/tensor_mapping.py

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

* fix tensor mapping

---------

Co-authored-by: Thireus ☠ <Thireus@users.noreply.github.com>
Co-authored-by: yairpatch <yairpatch@users.noreply.github.com>
Co-authored-by: LETS-BEE <LETS-BEE@users.noreply.github.com>
Co-authored-by: Xuan Son Nguyen <son@huggingface.co>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
This commit is contained in:
JJJYmmm
2025-10-30 16:19:14 +01:00
committed by GitHub
co-authored by Thireus ☠ yairpatch LETS-BEE Sigbjørn Skjæret Xuan Son Nguyen Georgi Gerganov
parent dcca0d3ab8
commit d261223d24
28 changed files with 1125 additions and 97 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/ggml-vulkan/ggml-vulkan.cpp
+4 -1
View File
@@ -1056,6 +1056,7 @@ struct vk_op_rope_push_constants {
uint32_t s1;
uint32_t s2;
int32_t sections[4];
uint32_t is_imrope;
uint32_t is_back;
uint32_t set_rows_stride;
};
@@ -9927,6 +9928,8 @@ static void ggml_vk_rope(ggml_backend_vk_context * ctx, vk_context& subctx, cons
memcpy(sections, (int32_t *) dst->op_params + 11, sizeof(int)*4);
}
const bool is_imrope = mode == GGML_ROPE_TYPE_IMROPE;
float corr_dims[2];
ggml_rope_yarn_corr_dims(n_dims, n_ctx_orig, freq_base, beta_fast, beta_slow, corr_dims);
@@ -9948,7 +9951,7 @@ static void ggml_vk_rope(ggml_backend_vk_context * ctx, vk_context& subctx, cons
(uint32_t)src0->ne[0], (uint32_t)n_dims, freq_scale, (uint32_t)src0->ne[1],
freq_base, ext_factor, attn_factor, {corr_dims[0], corr_dims[1]}, theta_scale,
src2 != nullptr, (uint32_t)src0->ne[2], s1, s2,
{ sections[0], sections[1], sections[2], sections[3] }, backprop, set_rows_stride,
{ sections[0], sections[1], sections[2], sections[3] }, is_imrope, backprop, set_rows_stride,
}, dryrun);
}
ggml/src/ggml-vulkan/vulkan-shaders/rope_head.glsl
+1
View File
@@ -27,6 +27,7 @@ layout (push_constant) uniform parameter {
uint s1;
uint s2;
int sections[4];
uint is_imrope;
uint is_back;
uint set_rows_stride;
} p;
ggml/src/ggml-vulkan/vulkan-shaders/rope_multi.comp
+23 -11
View File
@@ -32,17 +32,29 @@ void main() {
const uint sector = (i0 / 2) % sect_dims;
float theta_base = 0.0;
if (sector < p.sections[0]) {
theta_base = data_pos[channel_x]*pow(p.theta_scale, i0/2.0f);
}
else if (sector >= p.sections[0] && sector < sec_w) {
theta_base = data_pos[channel_x + ne2 * 1]*pow(p.theta_scale, i0/2.0f);
}
else if (sector >= sec_w && sector < sec_w + p.sections[2]) {
theta_base = data_pos[channel_x + ne2 * 2]*pow(p.theta_scale, i0/2.0f);
}
else if (sector >= sec_w + p.sections[2]) {
theta_base = data_pos[channel_x + ne2 * 3]*pow(p.theta_scale, i0/2.0f);
if (p.is_imrope != 0) {
if (sector % 3 == 1 && sector < 3 * p.sections[1]) {
theta_base = data_pos[channel_x + ne2 * 1]*pow(p.theta_scale, i0/2.0f);
} else if (sector % 3 == 2 && sector < 3 * p.sections[2]) {
theta_base = data_pos[channel_x + ne2 * 2]*pow(p.theta_scale, i0/2.0f);
} else if (sector % 3 == 0 && sector < 3 * p.sections[0]) {
theta_base = data_pos[channel_x]*pow(p.theta_scale, i0/2.0f);
} else {
theta_base = data_pos[channel_x + ne2 * 3]*pow(p.theta_scale, i0/2.0f);
}
} else {
if (sector < p.sections[0]) {
theta_base = data_pos[channel_x]*pow(p.theta_scale, i0/2.0f);
}
else if (sector >= p.sections[0] && sector < sec_w) {
theta_base = data_pos[channel_x + ne2 * 1]*pow(p.theta_scale, i0/2.0f);
}
else if (sector >= sec_w && sector < sec_w + p.sections[2]) {
theta_base = data_pos[channel_x + ne2 * 2]*pow(p.theta_scale, i0/2.0f);
}
else if (sector >= sec_w + p.sections[2]) {
theta_base = data_pos[channel_x + ne2 * 3]*pow(p.theta_scale, i0/2.0f);
}
}
const float freq_factor = p.has_ff != 0 ? data_ff[i0/2] : 1.0f;