mirrors/ollama-for-amd
1
0
Fork 0
mirror of https://github.com/likelovewant/ollama-for-amd.git synced 2025-12-21 14:26:30 +00:00
Code Issues Packages Projects Releases Wiki Activity

ggml: Increase maximum graph size

Browse Source 
The initial implementation of qwen3-vl:235b exceeded the maximum graph
size based on the number of tensors. Although this was later fixed
through the use of the mrope operation, we are close to the limit in
some cases. This updates to track the current llama.cpp usage of GGML.
This commit is contained in:
Jesse Gross
2025-10-30 14:30:31 -07:00
committed by Jesse Gross
parent d2158ca6f4
commit ef549d513c
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
ml/backend/ggml/ggml.go
View File

@@ -378,7 +378,7 @@ func New(modelPath string, params ml.BackendParams) (ml.Backend, error) {
}
}
maxGraphNodes := max(8192, len(meta.Tensors().Items())*5)
maxGraphNodes := max(1024, len(meta.Tensors().Items())*8)
sched := C.ggml_backend_sched_new_ext(
(*C.ggml_backend_t)(unsafe.Pointer(&schedBackends[0])),