Back to Freedom.Tech Back
All llama.cpp releasesAll versions
Release Thu, Aug 6, 2026 1 min read

llama.cpp b10290

Original release notes

mtmd/ggml: add ggml_build_forward_order (#26649)

  • ggml: add ggml_build_forward_order

ggml_build_forward_expand marks the tensor and all its ancestors for compute, so using it as a pure ordering hint (keeping q, k and v together) defeats ggml_build_forward_select: the unselected branch is forced to run with inputs that were never uploaded. In the mtmd audio graph this makes GEN_WAV calls execute the GEN_CODE branch with a stale inp_code0, hitting the get_rows bound assert on CPU.

Add ggml_build_forward_order, which inserts nodes without the compute flag; the flag is restored when the branch is actually selected. Switch the q/k/v hints in clip_graph::build_attn to it.

  • nit: reduce comments (AGENTS.md)

Website: -

UI: