CUDA: fix negative KV_max values in FA (llama/15321)

HIP: Cleanup hipification header (llama/15285)

cuda : fix GGML_CUDA_GRAPHS=OFF (llama/15300)

finetune: SGD optimizer, more CLI args (llama/13873)

HIP: bump requirement to rocm 6.1 (llama/15296)

CUDA: Optimize `reduce_rows_f32` kernel, leading up to 25x perf improvement on kernel-level and 10% perf increase for Gemma3n (llama/15132)

HIP: disable sync warp shuffel operators from clr amd_warp_sync_functions.h (llama/15273)

CUDA cmake: add `-lineinfo` for easier debug (llama/15260)

musa: fix failures in test-backend-ops for mul_mat_id op (llama/15236)

cuda: refactored ssm_scan and use CUB (llama/13291)

CUDA: add attention sinks for tile and wmma (llama/15178)

ggml : fix field name when new ggml_backend (llama/14944)

CUDA: attention sinks for mma FlashAttention (llama/15157)

CUDA: GEMM for FP32/FP16/BF16 and ne11 <= 16 (llama/15131)

llama : add gpt-oss (llama/15091)

CUDA: use mma FA kernel for gqa > 4 on RTX 4000 (llama/15035)

cuda: make im2col a little faster (llama/15025)

cuda, sycl : fix batched gemm when ne02 == 1 && ne03 > 1 (llama/15038)

CUDA: fix MMQ nwarps for AMD with warp_size==32 (llama/15014)

HIP: enable mfma mmq on gfx908 and gfx90a for select datatypes and shapes (llama/14949)

CUDA: skip masked KV slices for all FA kernels (llama/14924)

HIP: remove the use of __HIP_PLATFORM_AMD__, explicitly support only AMD targets (llama/14945)

HIP: add GGML_HIP_MMQ_MFMA option to allow disableing the MFMA path. (llama/14930)

HIP: Ignore unsupported unroll transformation in fattn-vec (llama/14931)

cuda : add softcap fusion (llama/14907)

CUDA: add roll (llama/14919)

CUDA: fix pointer incrementation in FA (llama/14916)

HIP: Enable Matrix cores for MMQ Kernels, Enable stream-K for CDNA 3 (llama/14624)

musa: fix build warnings (unused variable) (llama/14869)

musa: upgrade musa sdk to rc4.2.0 (llama/14498)

CUDA: fix overflow in FA, tune performance (llama/14840)

CUDA: fix compilation with GGML_CUDA_F16 (llama/14837)

CUDA: fix quantized KV cache + multiple sequences (llama/14822)

CUDA: add fused rms norm (llama/14800)

cuda : implement bf16 cpy ops and enable bf16 cont (llama/14763)

cuda: remove linking to cublasLt (llama/14790)

vulkan/cuda: Fix im2col when KW!=KH (llama/14789)

cuda : Fix Gemma3n not executed as CUDA_GRAPH on NVGPUs (llama/14741)

CUDA: set_rows + cpy.cu refactor (llama/14712)

llama : add high-throughput mode (llama/14363)

cuda: fix build warnings in set-rows.cu (unused variable) (llama/14687)

cuda : add set rows for bf16 (llama/14664)

cuda : add ELU support (llama/14657)

ggml : add build-time message to remind about ggml_set_rows (llama/14661)

CUDA: add set rows for f32 and f16 (llama/14551)

model : support LiquidAI LFM2 hybrid family (llama/14620)

HIP : Add HIP 7.0+ compatibility for hipBLAS compute types (llama/14634)

cuda : support Falcon-H1 state size for SSM_SCAN (llama/14602)

ggml : add ggml_scale_bias (llama/14417)

cuda : fix rope with partial rotation and non-cont src (llama/14580)