master
ggml: backend-agnostic tensor parallelism (experimental) (#19378)
ggml-cpu: Optimized x86 and generic cpu q1_0 dot (follow up) (#21636)
ggml : build backends as libraries (#10256)
kleidiai : fix MUL_MAT support for batched (3D) inputs (#20620)
ggml-cpu: fix fallback for RVV kernels without zvfh (#21157)
ggml : fix SpaceMit IME array out-of-bounds in task assignment (#16629)
fix(ggml): correct RISC-V ISA string canonical ordering for RVV in CMake (#20888)
ggml : extend bin bcast for permuted src1 (#19484)
cpu: de-duplicate some of the operators and refactor (ggml/1144)
ggml-cpu: FA add GEMM microkernel (#19422)
ggml : fix ARM NEON nvfp4 dot product on non-dotprod targets (#21559)
ggml: add Q1_0 1-bit quantization support (CPU) (#21273)
ggml-cpu : split arch-specific implementations (#13892)
ggml : fix a few instances of missing GGML_TYPE_Q1_0 cases (#21716)
ggml: add GATED_DELTA_NET op (#19504)
ggml-cpu: fix RVV checks in quants and repacking (#20682)
ggml-cpu: add RVV repack GEMM and GEMV for quantization types (#19121)
ggml : implemented simd_gemm kernel for riscv vector extension (#20627)
ggml : add native AVX512-FP16 support for F16 operations (#20529)
ggml : fix fallback to CPU for ununsupported ops (#15118)
ggml : unary ops support non-cont src0 + metal F16 unary ops (#19511)
ggml : add ops SOFTPLUS, EXPM1, TRI, SOLVE_TRI, CUMSUM (#17063)
ggml-cpu: optimize ggml_vec_dot_bf16 for s390x (#19399)