thek0tyara/llama-cpp-turboquant
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 4

model: LFM2-VL fixes (#17577)

Browse source 
* Adjust to pytorch

* Add antialiasing upscale

* Increase number of patches to 1024

* Handle default marker insertion for LFM2

* Switch to flag

* Reformat

* Cuda implementation of antialias kernel

* Change placement in ops.cpp

* consistent float literals

* Pad only for LFM2

* Address PR feedback

* Rollback default marker placement changes

* Fallback to CPU implementation for antialias implementation of upscale
This commit is contained in:
Tarek Dakhran 2025-11-30 21:57:31 +01:00 • committed by GitHub
parent 7f8ef50cce
commit 2ba719519d
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
12 changed files with 162 additions and 13 deletions
Download patch file Download diff file Expand all files Collapse all files

2
tests/test-backend-ops.cpp
View file

@ -7660,7 +7660,7 @@ static std::vector<std::unique_ptr<test_case>> make_test_cases_eval() {
// test_cases.emplace_back(new test_top_k(GGML_TYPE_F32, {i, 2, 1, 3}, rand() % i + 1));
//}
for (ggml_scale_mode mode : {GGML_SCALE_MODE_NEAREST, GGML_SCALE_MODE_BILINEAR, GGML_SCALE_MODE_BICUBIC}) {
for (ggml_scale_mode mode : {GGML_SCALE_MODE_NEAREST, GGML_SCALE_MODE_BILINEAR, GGML_SCALE_MODE_BICUBIC, ggml_scale_mode(GGML_SCALE_MODE_BILINEAR | GGML_SCALE_FLAG_ANTIALIAS)}) {
test_cases.emplace_back(new test_upscale(GGML_TYPE_F32, {512, 512, 3, 2}, 2, mode));
test_cases.emplace_back(new test_upscale(GGML_TYPE_F32, {512, 512, 3, 2}, 2, mode, true));
test_cases.emplace_back(new test_interpolate(GGML_TYPE_F32, {2, 5, 7, 11}, {5, 7, 11, 13}, mode));