Starstruk Logo MarkStarstruk

b9876

ggml-org/llama.cppb9876Jul 5, 2026by github-actions[bot]

AI Summary

Fixes a crash when using tensor parallelism (-sm tensor) combined with CPU-offloaded MoE experts (-ncmoe). The fix moves the split-state lookup before the contiguity assert.

Key Highlights

  • Fixes crash for tensor parallelism with CPU-offloaded MoE experts
  • Fixes GGML_ASSERT failure for non-contiguous mirrored tensors
  • Moves split-state lookup before contiguity check
  • Fixes issue #24886

New Features

  • Stability fix for tensor parallelism with MoE experts

Full Release Notes

<details open>

ggml : fix tensor-parallel + -ncmoe crash on MoE models (#25028)

Tensor parallelism (-sm tensor) combined with -ncmoe (CPU-offloaded MoE
experts) aborts during warm-up on MoE models with
GGML_ASSERT(ggml_is_contiguous(tensor)) in ggml-backend-meta.cpp.

The failing tensor is the MoE router output (ffn_moe_topk): it is mirrored
(GGML_BACKEND_SPLIT_AXIS_MIRRORED, replicated across backends since routing
must be identical) and happens to be a non-contiguous view.
ggml_backend_meta_buffer_{get,set}_tensor asserted contiguity before
consulting the split state, so a mirrored non-contiguous tensor tripped the
assert even though the GGML_BACKEND_SPLIT_AXIS_MIRRORED case right below
already handles it.

Move the split-state lookup above the assert and allow the mirrored case in
both get_tensor and set_tensor.

Diagnosis credit to the reporter (@nathanmp).

Fixes #24886

Signed-off-by: liminfei-amd <91481003+liminfei-amd@users.noreply.github.com>

</details>

**macOS/iOS:**
- [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/download/b9876/llama-b9876-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/b9876/llama-b9876-bin-macos-x64.tar.gz)
- [iOS XCFramework](https://github.com/ggml-org/llama.cpp/releases/download/b9876/llama-b9876-xcframework.zip)

**Linux:**
- [Ubuntu x64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b9876/llama-b9876-bin-ubuntu-x64.tar.gz)
- [Ubuntu arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b9876/llama-b9876-bin-ubuntu-arm64.tar.gz)
- [Ubuntu s390x (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b9876/llama-b9876-bin-ubuntu-s390x.tar.gz)
- [Ubuntu x64 (Vulkan)](https://github.com/ggml-org/llama.cpp/releases/download/b9876/llama-b9876-bin-ubuntu-vulkan-x64.tar.gz)
- [Ubuntu arm64 (Vulkan)](https://github.com/ggml-org/llama.cpp/releases/download/b9876/llama-b9876-bin-ubuntu-vulkan-arm64.tar.gz)
- [Ubuntu x64 (ROCm 7.2)](https://github.com/ggml-org/llama.cpp/releases/download/b9876/llama-b9876-bin-ubuntu-rocm-7.2-x64.tar.gz)
- [Ubuntu x64 (OpenVINO)](https://github.com/ggml-org/llama.cpp/releases/download/b9876/llama-b9876-bin-ubuntu-openvino-2026.2.1-x64.tar.gz)
- [Ubuntu x64 (SYCL FP32)](https://github.com/ggml-org/llama.cpp/releases/download/b9876/llama-b9876-bin-ubuntu-sycl-fp32-x64.tar.gz)
- [Ubuntu x64 (SYCL FP16)](https://github.com/ggml-org/llama.cpp/releases/download/b9876/llama-b9876-bin-ubuntu-sycl-fp16-x64.tar.gz)

**Android:**
- [Android arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b9876/llama-b9876-bin-android-arm64.tar.gz)

**Windows:**
- [Windows x64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b9876/llama-b9876-bin-win-cpu-x64.zip)
- [Windows arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b9876/llama-b9876-bin-win-cpu-arm64.zip)
- [Windows arm64 (OpenCL Adreno)](https://github.com/ggml-org/llama.cpp/releases/download/b9876/llama-b9876-bin-win-opencl-adreno-arm64.zip)
- [Windows x64 (CUDA 12)](https://github.com/ggml-org/llama.cpp/releases/download/b9876/llama-b9876-bin-win-cuda-12.4-x64.zip) - [CUDA 12.4 DLLs](https://github.com/ggml-org/llama.cpp/releases/download/b9876/cudart-llama-bin-win-cuda-12.4-x64.zip)
- [Windows x64 (CUDA 13)](https://github.com/ggml-org/llama.cpp/releases/download/b9876/llama-b9876-bin-win-cuda-13.3-x64.zip) - [CUDA 13.3 DLLs](https://github.com/ggml-org/llama.cpp/releases/download/b9876/cudart-llama-bin-win-cuda-13.3-x64.zip)
- [Windows x64 (Vulkan)](https://github.com/ggml-org/llama.cpp/releases/download/b9876/llama-b9876-bin-win-vulkan-x64.zip)
- [Windows x64 (OpenVINO)](https://github.com/ggml-org/llama.cpp/releases/download/b9876/llama-b9876-bin-win-openvino-2026.2.1-x64.zip)
- [Windows x64 (SYCL)](https://github.com/ggml-org/llama.cpp/releases/download/b9876/llama-b9876-bin-win-sycl-x64.zip)
- [Windows x64 (HIP)](https://github.com/ggml-org/llama.cpp/releases/download/b9876/llama-b9876-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/b9876/llama-b9876-ui.tar.gz)