b10290

ggml-org/llama.cppb10290Aug 6, 2026by github-actions[bot]

AI Summary

This release introduces a new graph ordering function (`ggml_build_forward_order`) to prevent stale inputs in multi-modal audio processing graphs. It addresses a bug where compute flags were applied too early, causing incorrect execution in specific branches.

Key Highlights

Added `ggml_build_forward_order` for better tensor ordering in multi-modal graphs.
Fixes stale input handling in GEN_WAV calls within mtmd audio graphs.
Prevents forced execution of unselected branches with stale inputs.
Updated comments in AGENTS.md.

Full Release Notes

<details open>

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)

</details>

**Website:**
- <https://llama.app>

**macOS/iOS:**
- [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/download/b10290/llama-b10290-bin-macos-arm64.tar.gz)
- macOS Apple Silicon (arm64, KleidiAI enabled) [DISABLED](https://github.com/ggml-org/llama.cpp/pull/23780)
- [macOS Intel (x64)](https://github.com/ggml-org/llama.cpp/releases/download/b10290/llama-b10290-bin-macos-x64.tar.gz)
- [iOS XCFramework](https://github.com/ggml-org/llama.cpp/releases/download/b10290/llama-b10290-xcframework.zip)

**Linux:**
- [Ubuntu x64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b10290/llama-b10290-bin-ubuntu-x64.tar.gz)
- [Ubuntu arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b10290/llama-b10290-bin-ubuntu-arm64.tar.gz)
- [Ubuntu s390x (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b10290/llama-b10290-bin-ubuntu-s390x.tar.gz)
- [Ubuntu x64 (Vulkan)](https://github.com/ggml-org/llama.cpp/releases/download/b10290/llama-b10290-bin-ubuntu-vulkan-x64.tar.gz)
- [Ubuntu arm64 (Vulkan)](https://github.com/ggml-org/llama.cpp/releases/download/b10290/llama-b10290-bin-ubuntu-vulkan-arm64.tar.gz)
- [Ubuntu x64 (ROCm 7.2)](https://github.com/ggml-org/llama.cpp/releases/download/b10290/llama-b10290-bin-ubuntu-rocm-7.2-x64.tar.gz)
- [Ubuntu x64 (OpenVINO)](https://github.com/ggml-org/llama.cpp/releases/download/b10290/llama-b10290-bin-ubuntu-openvino-2026.2.1-x64.tar.gz)
- [Ubuntu x64 (SYCL FP32)](https://github.com/ggml-org/llama.cpp/releases/download/b10290/llama-b10290-bin-ubuntu-sycl-fp32-x64.tar.gz)
- [Ubuntu x64 (SYCL FP16)](https://github.com/ggml-org/llama.cpp/releases/download/b10290/llama-b10290-bin-ubuntu-sycl-fp16-x64.tar.gz)

**Android:**
- [Android arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b10290/llama-b10290-bin-android-arm64.tar.gz)

**Windows:**
- [Windows x64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b10290/llama-b10290-bin-win-cpu-x64.zip)
- [Windows arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b10290/llama-b10290-bin-win-cpu-arm64.zip)
- [Windows arm64 (OpenCL Adreno)](https://github.com/ggml-org/llama.cpp/releases/download/b10290/llama-b10290-bin-win-opencl-adreno-arm64.zip)
- [Windows x64 (CUDA 12)](https://github.com/ggml-org/llama.cpp/releases/download/b10290/llama-b10290-bin-win-cuda-12.4-x64.zip) - [CUDA 12.4 DLLs](https://github.com/ggml-org/llama.cpp/releases/download/b10290/cudart-llama-bin-win-cuda-12.4-x64.zip)
- [Windows x64 (CUDA 13)](https://github.com/ggml-org/llama.cpp/releases/download/b10290/llama-b10290-bin-win-cuda-13.3-x64.zip) - [CUDA 13.3 DLLs](https://github.com/ggml-org/llama.cpp/releases/download/b10290/cudart-llama-bin-win-cuda-13.3-x64.zip)
- [Windows x64 (Vulkan)](https://github.com/ggml-org/llama.cpp/releases/download/b10290/llama-b10290-bin-win-vulkan-x64.zip)
- [Windows x64 (OpenVINO)](https://github.com/ggml-org/llama.cpp/releases/download/b10290/llama-b10290-bin-win-openvino-2026.2.1-x64.zip)
- [Windows x64 (SYCL)](https://github.com/ggml-org/llama.cpp/releases/download/b10290/llama-b10290-bin-win-sycl-x64.zip)
- [Windows x64 (HIP)](https://github.com/ggml-org/llama.cpp/releases/download/b10290/llama-b10290-bin-win-hip-radeon-x64.zip)

**openEuler:**
- [DISABLED](https://github.com/ggml-org/llama.cpp/pull/23705)
- openEuler x86 (310p)
- openEuler x86 (910b, ACL Graph)
- openEuler aarch64 (310p)
- openEuler aarch64 (910b, ACL Graph)

**UI:**
- [UI](https://github.com/ggml-org/llama.cpp/releases/download/b10290/llama-b10290-ui.tar.gz)