Nightly build: b4950

Change log since v1.8.0

161c968 Revert "ci : comment out all but one build in release.yml"
66101c7 ci : remove unused ccache step
20aef41 cmake : add WHISPER_USE_SYSTEM_LLAMA option [no ci]
858ad88 ci : add attestation for signed release artifacts
d821db8 ci : add missing GGML_NATIVE=OFF to jobs
ee00221 ci: add GGML_NATIVE=OFF to build-clang.yml
7fc8845 examples : use FetchContent for llama.cpp in talk-llama
452005a fix move artifacts step
7715478 ci : add github rulesets for releases
33ebdd4 ci : use DEPLOY_KEY_RELEASE
24b5677 ci : comment out all but one build in release.yml
127af86 ci : fix if statement in release.yml
43e5d9f ci : update release workflow to be consistent with llama.cpp
b6e63fa cmake : update semver handling to be consistent with ggml/llama.cpp
4834a23 talk-llama : sync llama.cpp
6b014cf sync : ggml
8189458 ggml : bump version to 0.20.2 (ggml/1589)
51319a2 CUDA: MMVQ nwarps=8 for bs=1 for dense models on DGX Spark (llama/26843)
71759b7 cuda : skip UMA override for HIP builds (llama/27083)
964bb1b ggml : bump version to 0.20.1 (ggml/1587)
4257f47 sycl: fix thread/block count in quantized cpy kernel launches (llama/27160)
0f38613 support OP OPT_STEP_ADAMW, OPT_STEP_SGD (llama/25268)
9a0d190 vulkan: add SHMEM_STRIDE_PAD/APPLY_SLM_A_RESHAPE for coopmat1 on Intel Xe (llama/25380)
ab71410 fixed indent
9b98e57 Fixed gating logic for problematic Intel driver version
1fe009c talk-llama : fix build (#0)
733f281 sync : ggml
667da04 ggml : bump version to 0.20.0 (ggml/1584)
b7ea8b1 ggml : recurrent state rollback for ggml_ssm_scan (llama/26623)
2aef2a0 sycl: fuse mul_mat(gate) + mul_mat(up) + GLU for q4_K dense FFN (llama/26779)
5bc0c85 ggml: force single thread on wasi (llama/25686)
2d05b6e sycl: fuse the gated-delta-net state writeback cpy (llama/26643)
43cbe41 OpenVINO: Qwen3.5, memory optimization, and test-recurrent-state-rollback (llama/26952)
94d2d35 Support host pinned mem to improve SYCL Host-to-Device Memory Access (llama/26789)
34009e8 metal : add TQ2_0 support (llama/26980)
62031fe ggml-cpu/ops: vectorize flash-attention V-cache F16 to F32 conversion (llama/26947)
ac9a74f sycl: remove separate fp32 type promotion in gemm non-oneDNN path (llama/26372)
3425d13 sycl: fuse UNARY(silu|sigmoid|softplus) + MUL (llama/26411)
90d4ed1 sycl : Add DMMV ESIMD Q3_K kernel (llama/26251)
f701160 sycl : enhance concat to support Q4_0, Q4_1, Q5_0, Q5_1, Q8_0 (llama/26800)
c7be8b3 ggml-hip : remove -funsafe-math-optimizations (llama/26696)
406b116 ggml : fix arm builds, unused var (llama/26991)
b0e3297 gguf : harden loader against malformed tensor dims and metadata types (llama/25596)
92423af kleidiai: Add runtime feature detection mechanism for aarch64/kleidiai (llama/26076)
1b93067 opencl: default FA c8 cluster width to 16 on X1E (llama/26433)
6329040 vulkan: add TQ2_0 (ternary) support (llama/25850)
db3687b opencl: use flat mv q5_k when weight exceeds image1d_buffer_t limit (llama/26880)
cdbe455 CUDA: only disable CUDA graphs when mul_mat_id actually needs a stream sync (llama/26802)
030656f cuda : add warp-per-row wkv7 kernel for single-token decode (llama/26111)
d958968 llama: add default load-mode auto, which avoids mmap on iGPUs (llama/26081)
0f60f8e ggml-webgpu: fix CI errors from #25025 and #25262 (llama/26566)
a6e2630 opencl: transpose the K tile in local memory for FA prefill kernels (llama/26428)
830ec22 ggml-cpu : fix CPU affinity mask being ignored on Android (llama/26838)
b3bc904 ggml : require contiguous src for ROLL on CUDA and Metal (llama/25928)
f09a97c ggml-webgpu : refactor several wgsl files and simplify flash_attn wgsl. (llama/26134)
877761c ggml-cpu : fix missing Q5_0 dispatch in SpaceMiT backend (llama/26792)
10791af CUDA: fuse rms_norm + mul + rope (+ view + set_rows) (llama/26767)
068d3b0 CUDA: fix thread/block count in quantized cpy kernel launches (llama/26731)
eb3296f sycl: coalesce the ssm_conv window loads (llama/26612)
8a5ba01 metal : fix NORM/RMS_NORM for row lengths that leave a partial simdgroup (llama/26708)
06248ca cmake : add config version support (ggml/1582)
592feef talk-llama : sync llama.cpp
8770492 sync : ggml
84cdcad ggml : bump version to 0.19.0 (ggml/1581)
8587ad3 ggml : add aarch64 HWCAP fallbacks and fix fp16 variant detection (llama/25554)
56cb154 sycl: fix UE4M3 parsing (llama/25608)
077c5d4 sycl: *glu flat path (llama/26354)
9faa9ee sycl : Support DSv4 OPs: LIGHTNING_INDEXER,DSV4_HC_COMB,DSV4_HC_POST,DSV4_HC_PRE (llama/26568)
fb9e8ca sycl : fix error Error OP FLASH_ATTN_EXT on arc770 (llama/26441)
89d45af sycl : enhance OP set_rows to support all missed data types (llama/26515)
79ab70c cuda: fix warnings for unused variable/function (llama/26688)
5693378 metal : avoid threadgroup matrix array instantiation in kernel_lightning_indexer (llama/26646)
69bd0a9 ci : onboard AMD ROCm CI with gfx1151 fixes (llama/26544)
5a80d0a vulkan: fix submission batching size, add debug tools for diagnosing causes of DeviceLost drivers errors (llama/26371)
60894f1 mtmd/ggml: add ggml_build_forward_order (llama/26649)
8731021 vulkan backend ops: implemented GATED_LINEAR_ATTN (llama/25601)
8631825 whisper : heap out-of-bounds read in log_mel_spectrogram on very short audio (ggml-org#3956)
df1547b whisper,parakeet : reject invalid n_dims in tensor header to prevent stack-buffer-overflow on malformed model files (ggml-org#3957)
306c88f release : v1.9.2 (ggml-org#3970)
6b92fec talk-llama : sync llama.cpp
8184446 sync : ggml
2a502dc ggml : bump version to 0.18.1 (ggml/1578)
6217ef7 sycl: parallelize the non-contiguous concat kernel (llama/25852)
97abf5c Extended SYCL oneDNN SDPA to non-FP16 KV caches (Q4_0–Q8_0 and FP32) (#25874)
b5dec64 ggml: use dynamic allocation for split graph inputs (llama/22789)
4673c4b opencl: route large q6_K lm_head to the flat GEMV (llama/26427)
3fee8a1 CUDA: Fix data-races when reusing SMEM in block_reduce (llama/26385)
e5f6c74 metal: implement DSv4 Lightning Indexer (llama/25893)
319e9a7 metal : add SILU_BACK (llama/25982)
4695615 metal : add F16 support for bin ops (llama/26465)
78f340b opencl: limit local workgroup size for GLU operation (llama/26383)
e7ae4d7 metal: implement DeepSeek V4 hyper-connections (llama/26459)
4e7439e opencl: bugfix increment ref_count in ggml_backend_opencl_init() (llama/26162)
d735ac6 sycl: fix classification of iGPUs (llama/26105)
cabd684 ggml-webgpu: add support for f16 repeat (llama/26307)
83105b7 vulkan: extend topk_moe fusion to support sqrt(softplus) (llama/26124)
cb62586 vulkan: add POOL_1D op (llama/25431)
153a336 vulkan: Introduce driver version check for Windows Intel GPU to mitigate crashing (llama/25192)
50c198c cuda: extract Q2_0 elements via __byte_perm (llama/25603)
748b84b SYCL: add oneMKL GEMM flash attention for XMX-accelerated prompt proc… (#25025)
67ed569 support the missed types in cpy (llama/26005)
a2c3a61 ggml-zendnn : group matmul direct API for mul_mat_id (llama/25918)
fcb5b86 sycl : support dev2dev memcpy by DEV2DEV_MEMCPY_FORWARD (llama/26234)
2dba496 Support q2 mul_mat (llama/26231)
466b016 sycl: fuse RMS_NORM + MUL (llama/26015)
f974175 ggml-webgpu: improve flash_attn_vec for quantized KV at long contexts (llama/25956)
23feefc vulkan: Support quantized concat (llama/25684)
1e64098 Test support for alternative conv layout (llama/25617)
1a8c5a6 ggml-cuda: Allow transpose-free gemmv computation (llama/26171)
64d57d3 ci : set GGML_NATIVE=OFF for build-vad workflow (ggml-org#3966)
2957983 examples : fix VAD min silence argument parsing (ggml-org#3963)
2ca53bb sync : ggml (ggml-org#3962)
4523d0c parakeet : verify hparams loaded from parakeet model bin file (ggml-org#3950)
a630b35 Adding ruby bindings change log (ggml-org#3959)
97c56f1 ruby : add VAD speech segments API (ggml-org#3931)
080bbbe examples : Remove leading space from txt output (ggml-org#3921)
7695a53 ggml : use ggml_vqtbl1q_u8 for 32-bit compat (#0)
289ecb0 talk-llama : sync llama.cpp
afa6bff sync : ggml
4fc2a2d ggml : bump version to 0.16.0 (ggml/1559)
6b17d2c ggml : fix conv 2d dw (llama/25490)
ade3292 common : adapt to q2_0 (ggml/0)
a2c5028 Only index by compile times + always multiply/add (llama/25445)
9cd788e metal : add CONV_2D_DW (depthwise convolution) support (llama/21565)
cc7a521 ggml-hip: enable -funsafe-math-optimizations (llama/24668)
1b0b078 cuda: align snake fusion matcher with the other backends (llama/25460)
def36a5 hexagon: add VISION RoPE support (llama/25216)
4b91409 ggml-webgpu: tune subgroup split (d_split) in flash_attn_vec (llama/25418)
8e7b672 opencl: Q6_K GEMM/GEMV fix for ne01 of weights that are not multiples of 128. (llama/25464)
b47f39d vulkan: disable FA mask_opt on GCN to improve performance (llama/24362)
5f052fb opencl: ragged-tile MoE prefill FP16 GEMM optimization (skip padded expert tiles) (llama/25433)
06492a1 vulkan: for small AMD GPUs, reduce submission threshold based on CU count (llama/25240)
50e0adb hexagon: new vtcm layouts and improved pipelines for MUL_MAT, MUL_MAT_ID and FLASH_ATTN_EXT (llama/25425)
b83d167 cuda : add support for f16->f16 GGML_OP_SET_ROWS (llama/25367)
81c8bb9 metal : add set_rows with src0 f16 (llama/25434)
3d521d1 ggml : fix A indexing in simd_gemm scalar tail-column path (llama/25390)
72e3f21 ggml : add support for CPU f16->f16 GGML_OP_SET_ROWS (llama/25344)
75f1f0c opencl: fix potential crash in aos reconstruct (llama/25383)
66e178e Add Q2_0 quantization: type definition and CPU backend (llama/24448)
2061e3b CUDA: Fuse MMVQ post-scale for NVFP4 (llama/24481)
38d16b9 ggml-hip : add -fno-finite-math-only alongside -ffast-math (llama/25373)
7c04837 fix unsupported UT cases of CONT & CPY (llama/25231)
bcbc5bc support op col2im_1d (llama/25264)
a0b8896 support OP cross_entropy_loss, cros...