tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-07-19 09:08:04 +00:00
Code Issues Packages Projects Releases Wiki Activity

llama : add Qwen2VL support + multimodal RoPE (#10361)

Browse Source 
* Barebone Qwen2VL LLM convertor

* Add Qwen2VL cli entrypoint

* [WIP] add qwen2vl arch

* Verify m-rope output

* Add vl-rope/2d-rope support for qwen2vl ViT

* update qwen2vl cli tool

* update 5D tensor op workaround

* [WIP] qwen2vl vision model

* make batch and clip utils compatible with qwen2vl

* [WIP] create inference workflow, gguf convert script but fix

* correcting vision-rope behavior, add the missing last layer back to ViT

* add arg parser to qwen2vl_surgery

* replace variable size array with vector

* cuda-gdb cmake preset

* add fp32 mrope, vision rope kernel

* add fp16 support for qwen2vl and m-rope

* add `GGML_ROPE_TYPE_MROPE`, `GGML_ROPE_TYPE_VISION`

* fix rope op mode switching, out dated func args

* update `llama_hparams`

* update to keep up stream changes

* resolve linter, test errors

* add makefile entry, update speical image padding token

* add mrope unit test, fix few compiler warnings

* rename `mrope` related function, params

* minor updates on debug util, bug fixs

* add `m-rope` testcase to `test-backend-ops`

* Apply suggestions from code review

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>

* fix traililng whitespce

* store `llama_hparams.rope_sections` with fixed size array

* update position id tensor size check in GGML_OP_ROPE

* minor updates

* update `ggml_backend_*_supports_op` of unsupported backends

* remote old `rope_section` compare operator

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
This commit is contained in:
HimariO
2024-12-14 20:43:46 +08:00
committed by GitHub
parent 56eea0781c
commit ba1cb19cdd
24 changed files with 1873 additions and 114 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
ggml/src/ggml-metal/ggml-metal.m
View File

@ -1125,8 +1125,18 @@ static bool ggml_metal_supports_op(const struct ggml_backend_metal_device_contex
return has_simdgroup_reduction && (op->ne[0] % 4 == 0);
case GGML_OP_ARGMAX:
case GGML_OP_NORM:
case GGML_OP_ROPE:
return true;
case GGML_OP_ROPE:
{
const int mode = ((const int32_t *) op->op_params)[2];
if (mode & GGML_ROPE_TYPE_MROPE) {
return false;
}
if (mode & GGML_ROPE_TYPE_VISION) {
return false;
}
return true;
}
case GGML_OP_IM2COL:
return op->src[0]->type == GGML_TYPE_F16;
case GGML_OP_POOL_1D:
@ -3026,7 +3036,9 @@ static void ggml_metal_encode_node(
} break;
case GGML_OP_ROPE:
{
GGML_ASSERT(ne10 == ne02);
// make sure we have one or more position id(ne10) per token(ne02)
GGML_ASSERT(ne10 % ne02 == 0);
GGML_ASSERT(ne10 >= ne02);
const int nth = MIN(1024, ne00);