EngineX-Ascend/enginex-ascend-910-llama.cpp
10
0
Fork 0
Code Issues Pull Requests Actions 4 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 23:19:14 +08:00
committed by GitHub
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

5
ggml/src/ggml-vulkan/ggml-vulkan.cpp
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);
}