Search headlines, summaries and articles to connect the dots.
Optimized TOP_K performance on Hexagon backend by skipping bitonic sort for tiles whose maximum cannot surpass the k-th best value, replacing tile sort with top_k_scan() to reduce 256-float blocks using HVX max-reduce, and inserting only surviving elements into a sorted k-entry list. Supports k values from 2 to 64 and row widths over 1024 in both single-row and multi-row paths. NaN treated as -INF, ties preserve first-seen index. Eliminated scalar reads from VTCM in hex-top-k.
The ggml library now checks view bounds based on the actual bytes reached by strides, improving accuracy for strided views. Views are required to start from contiguous elements along dimension 0, ensuring correct data access. Convolution windows in delta_net are created with all rows at once to prevent stride-related issues.
Dify 1.19.0 enhances observability with nested workflow tracing, allowing users to follow execution steps inside Workflow Tools, inspect LLM calls, and maintain context across nested invocations. The release also enables portable app packages via App packages, improves Agent file uploads and previews, introduces Skill Hub automation through CLI commands, and upgrades the bundled plugin daemon to version 0.6.14-local. Key changes include support for format v2 app packages, enhanced security, improved editing workflows, and updated environment variables and API endpoints.
kv-cache : never return 0 kv_rows for sequences without cells (#31570) A cache that shares the cells of another one (f.ex. the gemma4 assistant) can build a graph before its sequences have any cells, f.ex. at the server warmup. get_n_kv_rows() then returned 0, which means that kv_rows are not used, so the graph fell back to the dense KQ mask and the scheduler re-planned for it. Later graphs with larger masks then aborted with GGML_SCHED_DEBUG_REALLOC=1. **Website:** - **Attestations:** - **macOS/iOS:** - [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/download/b12065/llama-b12065-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/l
ggml : relax scale to non-contiguous views (#31146) * ggml : relax scale to non-contiguous views SCALE requires src0 to be 1D-contiguous. Relax this to dim-0 contiguity with arbitrary row strides (ggml_is_contiguous_rows), same as the unary ops in the Metal backend. - ggml: relax the assert in ggml_scale_impl - cpu: address src0 rows through their byte strides (f32, bf16) - cuda: stride-aware kernel, fastdiv row decomposition - metal: support bf16 scale (guarded unary instantiations) - sycl: keep the gate at contiguous, the kernel is flat f32 Add view variants to test_scale. Verified on CPU, Metal and CUDA. Assisted-by: dsh:llama.cpp/ggml-org/Qwen3.8-Flash-Next-GGUF (RTX 5090 32GB, DDR5 128GB) * cuda : skip the softcap fusion on non-contiguous views The fused softcap kernel reads src and w
tests : fix multi-seq split replay and partial-state cases for trained models (#31216) * tests : fix multi-seq split replay against trained models The rollback snapshot is captured by the graph that decoded the rolled-back tail, while the reference context decodes the same prefix with different batch shapes. The two states differ in the last bits and trained models amplify the difference with every replayed token. Compare only the first replayed token of each seq, then restore the rollback state into the reference context before the seq-1-only tail. Assisted-by: dsh:llama.cpp/ggml-org/Qwen3.8-Flash-Next-GGUF (MacBook, M5 Max, 128GB) * tests : require bit-exact seq-1 isolation in split replay The seq state save is not always bit-lossless - SWA-masked cells are not saved - so reload the rest
ggml-cpu : rewrite concat as one row kernel for all types (#31179) Every dst row comes from src0 or src1, or for dim 0 from the row of src0 followed by the row of src1, so a single kernel covers every type: a dense source row is one memcpy, quantized types included, and a strided source is copied element by element with the element size known at compile time. Rows are split across threads instead of i2 alone, which is often 1, and are copied in tiles of 32: for a transposed source the elements are walked in the outer loop, so the rows of a tile read neighboring addresses instead of jumping a full row stride per element. Add transposed b cases to test_concat, the layout of the delta-net conv input, and a perf case for it. **Website:** - **Attestations:** - **macOS/iOS:** - [macOS Apple Sili
llama-quant : fix 4D tensor quantization using ggml_nrows() and name-based expert detection (#30820) * llama-quant : fix quantizing 4D tensors nrows_total was ne[1]*ne[2], so for a 4D tensor only the first ne[2] slices were quantized and the rest of the rows were dropped. Use ggml_nrows(tensor) to count all rows, including ne[3]. Fixes #29916 * llama-quant : detect expert tensors by name and fix imatrix size for 4D tensors The MXFP4_MOE path used ne[2] > 1 to detect MoE expert tensors. This is wrong for non-expert tensors with 3 or more dims, e.g. MLA projections (BailingMoE3) and 4D conv kernels. Use the tensor name instead: all tensors used by GGML_OP_MUL_MAT_ID are named *_exps or *_chexps. This also removes the BailingMoE3 special case. The imatrix holds one vector of ne[0] values per
CUDA: fuse FFN gate/up matmuls and GLU in MMQ with interleaved gate/up rows (#29948) * cuda: fuse FFN gate/up and GLU in MMQ with interleaved gate/up rows For prompt processing, run the gate and up matmuls of a dense FFN as one mul_mat_q launch and apply the GLU in the write-back. - The loaders take the gate weights as a second pointer and fill the weight tile with up and gate rows interleaved in chunks of 8, so each thread holds an up value and the matching gate value (rows r and r + 8 of an MMA tile). - The MMA loop is unchanged: same tile, warp layout, vec_dot and one sum[] per thread. The fused write-back computes the GLU from registers and writes half the rows. - The fused path uses the regular tiling (no stream-k) with the column tiles innermost so that neighbouring blocks share weig
ggml-cuda: fuse qwen MRoPE + set-rows (#31025) **Website:** - **Attestations:** - **macOS/iOS:** - [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/download/b11878/llama-b11878-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/b11878/llama-b11878-bin-macos-x64.tar.gz) - [iOS XCFramework](https://github.com/ggml-org/llama.cpp/releases/download/b11878/llama-b11878-xcframework.zip) **Linux:** - [Ubuntu x64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b11878/llama-b11878-bin-ubuntu-x64.tar.gz) - [Ubuntu arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b11878/llama-b11878
