vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2026-05-11 04:51:25 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Merge branch 'upstream' into concedo_experimental

Browse source 
# Conflicts:
#	examples/model-conversion/Makefile
#	examples/model-conversion/README.md
#	examples/model-conversion/scripts/utils/quantize.sh
#	ggml/src/ggml-cpu/CMakeLists.txt
#	ggml/src/ggml-opencl/ggml-opencl.cpp
#	ggml/src/ggml-opencl/kernels/group_norm.cl
#	ggml/src/ggml-opencl/kernels/norm.cl
#	ggml/src/ggml-sycl/ggml-sycl.cpp
#	tests/test-backend-ops.cpp
#	tests/test-opt.cpp
#	tools/batched-bench/batched-bench.cpp
#	tools/mtmd/CMakeLists.txt
This commit is contained in:
Concedo 2025-08-27 17:39:24 +08:00
commit 654b9eee73
11 changed files with 595 additions and 178 deletions
Download patch file Download diff file Expand all files Collapse all files

2
src/llama-graph.cpp
View file

@ -1376,7 +1376,7 @@ ggml_tensor * llm_graph_context::build_attn(
// [TAG_NO_CACHE_PAD]
// TODO: if ubatch.equal_seqs() == true, we can split the three tensors below into ubatch.n_seqs_unq streams
assert(!ubatch.equal_seqs());
assert(!ubatch.equal_seqs() || (k_cur->ne[3] == 1 && k_cur->ne[3] == ubatch.n_seqs_unq));
ggml_tensor * q = q_cur;
ggml_tensor * k = k_cur;