Search: “llama-cpp-releases”

metal : add sparse FA (#28098) metal : support n_kv_max sparse mask hint in flash attention vec kernel add kernel_flash_attn_ext_vec_idx: compacts finite mask entries into a per-row index list (Hillis

mtmd: propagate const to preproc class (#28310) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/44990064 macOS/iOS: macOS Apple Silicon (arm64) macOS Apple

tune MMVQ to MMQ crossover for SM87 (#28285) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/45087167 macOS/iOS: macOS Apple Silicon (arm64) macOS Apple Sil

opencl: quant lm_head / decode GEMV and medium-batch GEMM optimizations (speculative decoding/MTP) (#26477) opencl: quant lm_head / decode GEMV and medium-batch GEMM optimizations opencl: guard q4_K/q

common/json-schema : fix GBNF grammar generation for empty object schemas (#28279) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/45102223 macOS/iOS: macOS

llama: fix whole source code rebuilt on each new commit (#28278) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/45114012 macOS/iOS: macOS Apple Silicon (ar

metal : add metallib build support for xcframework (#28163) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/44563342 macOS/iOS: macOS Apple Silicon (arm64)

hexagon: add missing FARF logs for cpy/get_rows/set_rows/gdn ops (#28217) hexagon: fix bug ne[2] printed in proc_op_req prep-src log hexagon: add shape/VTCM farf logs to cpy, get/set rows, gdn Website

opencl: fix out‐of‐bound reads in the Adreno image kernels (#27632) opencl: clamp the q4_K decode GEMV's fetch row on a padded x-grid opencl: enforce the tiling contract of the image KQ/KQV GEMMs open

vulkan : only request VK_KHR_shader_bfloat16 extension if supported (#28155) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/44637403 macOS/iOS: macOS Apple

vulkan: handle larger batch sizes (>4) efficiently for IQ3_S mat-vec (#27449) vulkan: handle larger batch sizes (>4) efficiently for IQ3_S mat-vec when NUM_COLS > 4. 5x perf at n=8 Assisted-by: Claude

hexagon: MUL_MAT and MUL_MAT_ID fusion and fixes (#28202) hex-mm: fuse QKV and FFN matmuls that land on HMX hex-mm: remove hardcoded ne[1] < 32K restriction hex-get-rows: explicitly reject repacked Q8

ggml: avoid KleidiAI buffer type init on dispatch (#27891) Co-authored-by: Acmmi acmmi@Acmmis-MacBook-Air.local Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestati

mtmd: Fix Qwen3-tts-0.6b (#28231) mtmd: load the qwen3-tts code predictor proj_in as optional The talker and the code predictor share the hidden size on the 0.6B checkpoints, so the reference builds n

mtmd: support DeepSeek-V4-Flash-Vision-Exp (#28133) mtmd: support DeepSeek-V4-Flash-Vision-Exp handle min/max token counts from CLI rm debugging use GGML_ROPE_TYPE_VISION nits apply review comments co

common, server : enable preserve_reasoning kwarg by default, log its effective state (#28174) common, server : enable preserve_reasoning kwarg by default, log its effective state If the preserve_reaso

ggml-cuda : remove unused vars (#28235) Signed-off-by: Adrien Gallouët angt@huggingface.co Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/44811308 macOS/iO

model: correctly support input vision for deepseek4 (#28154) model: correctly support input vision for deepseek4 nits Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/att

Update ROCm to 10.0.0 release (#27803) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/44848715 macOS/iOS: macOS Apple Silicon (arm64) macOS Apple Silicon (

metal : fix memory query under low-memory conditions (#27701) metal: Fix memory query under low-memory conditions Simply variable name Co-authored-by: Georgi Gerganov ggerganov@gmail.com Write it even

qwen4exp: sum the indexer heads by slices (#28023) qwen4exp: sum the indexer heads by slices The head reduction went through a transpose and a sum_rows over ne[1], which left sum_rows with ne0 = 4, on

qwen4exp: support recurrent state rollback (#28123) MTP speculative decoding needs the target state to move back by the number of rejected draft tokens. Without rollback support the context is classif

ci: Bump ggml-org/ccache-action to v1.2.24 (#28083) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/44375174 macOS/iOS: macOS Apple Silicon (arm64) macOS Ap

metal: enable Metal 4.0 tensor API on M5+/A19+ (#27461) metal : request Metal 4.0 language version for the tensor API metal : load the tensor API kernels from a separate metallib tests : add external-

tests : fix log verbosity for test-llama-archs (#28147) tests : fix log verbosity for test-llama-archs cont : naming cont : add note Website: https://llama.app Attestations: https://github.com/ggml-or

qwen4exp: fix seq_cp, block position keying, mtmd input, cuda abort, add tests (#27941) qwen4exp: follow up fixes -kvu NaN collapse fix Assisted-by: Claude indexer cache ext.x/ext.y restore fix Assist

sycl : support limit max alloc memory within 2GB for host-pinned memory (#27559) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/44438607 macOS/iOS: macOS A

metal : add fa-vec tuning for M2 Max (#28015) Rows for M2 Max (30 GPU cores) collected with 'ggml-metal-tuning fa-vec --dtype f16,q8_0', pasted into fa_vec_tuned_table. ref: https://github.com/ggml-or

metal : fix more leaks due to missing autoreleasepools (#27883) metal : fix more leaks due to missing autoreleasepools metal : rename variable metal : fix another missing pool warning Co-authored-by:

model : load hparams.n_layer_nextn before n_layer() calls (#28159) load hparams.n_layer_nextn before n_layer() calls remove duplicate loads Website: https://llama.app Attestations: https://github.com/

metal : add fa-vec tunings for A18 Pro (MacBook Neo) (#28152) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/44474754 macOS/iOS: macOS Apple Silicon (arm64

metal : add fa-vec tuning for M2 Pro (#28122) metal: add fa-vec tuning for M2 Pro metal : update fa-vec tuning for M2 Pro with new dtypes Website: https://llama.app Attestations: https://github.com/gg

context : autoscale n_ctx_train when yarn scaling specified (#28030) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/44499127 macOS/iOS: macOS Apple Silicon

kv-cells: look up the n-gram history in the sequence position index (#28040) get_prev_tokens() rebuilt a (seq, pos) -> token hash map on every ubatch by walking all used cells, while llama_kv_cells al

cuda: fuse MoE weighted expert reduction (#25952) cuda : fuse MoE weighted reduction (mul + view + add) The MoE combine tail currently writes weighted expert outputs to global memory before reducing t

ggml: add SWIGLU_CLAMP (#27930) ggml: add SWIGLU_CLAMP add vulkan shader Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/44034152 macOS/iOS: macOS Apple Sil

kv-cells: stop the sequence scan once all sequences are seen (#28011) for_each_token_in tested all LLAMA_MAX_SEQ sequences for every used cell, while a cell almost always belongs to one. The scan now

ggml : fix ggml_backend_buft_get_alloc_size() guard (#28038) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/44038725 macOS/iOS: macOS Apple Silicon (arm64)

rpc: avoid serializing buffers from other servers (#26500) rpc: avoid serializing buffers from other servers Only include remote buffer pointers when the buffer belongs to the RPC dispatcher receiving

metal : add remaining Q4_1/Q5_0/Q5_1 fa-vec tunings for M2 (#28017) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/44044108 macOS/iOS: macOS Apple Silicon

hexagon: fix CPY fence bug (#28033) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/44046740 macOS/iOS: macOS Apple Silicon (arm64) macOS Apple Silicon (arm

vulkan: top_k radix select for k >= 1024 for Qwen 3.8 Flash Next (#28032) vulkan: add top-k radix sort shader for k >= 1024 add Qwen 3.8 Flash Next top-k tests add top-k qsa fusion clean up code Websi

ggml : add MUL_MAT to the list of ops that may need additional memory (for WebGPU) (#28071) Co-authored-by: Stanisław Szymczyk sszymczy@gmail.com Website: https://llama.app Attestations: https://githu

vulkan: tune mat-vec rows for batched inference on Strix Halo (#27909) vulkan: RDNA3 static mat-vec rows above four columns On RDNA3 above four columns a static 4 rows for all types benches faster tha

spec : fuse the DFlash encoder into the KV cache injection (#27310) dflash : fuse the encoder into the KV injection decode The encoder is a single fc + norm, but running it as a separate llama_encode

sycl : Enhance to get the free memory of Intel GPU (#27968) enhance get mem info by l0 an SYCL API remove debug code, format the code update SYCL.md for GGML_SYCL_GET_MEM_API Website: https://llama.ap