b11321
BLAS : Document AOCL-BLAS build and label the device AOCL-BLAS (#29640) BLAS : Document AOCL-BLAS build and label the device AOCL-BLAS AOCL-Blas : Add an AOCL-BLAS Quick Start and drop the fixed versi
Knowledge catalogue
BLAS : Document AOCL-BLAS build and label the device AOCL-BLAS (#29640) BLAS : Document AOCL-BLAS build and label the device AOCL-BLAS AOCL-Blas : Add an AOCL-BLAS Quick Start and drop the fixed versi
hex-workqueue: fix race condition in seqn getting out of sync with idx_read/write (#29785) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/51906249 macOS/iO
HIP: avoid treating CDNA as dgx spark for gqa_ratio 20 in fattn_mma dqk 576 (#29572) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/51919562 macOS/iOS: mac
llama-mmap : avoid a second full-size copy of each tensor with direct-io (#29749) Assisted-by: Claude Co-authored-by: Pranesh Gonegandla pgonegandla@nvidia.com Website: https://llama.app Attestations:
arXiv:2609.38294v1 Announce Type: new Abstract: We study the generation of agentic workflows that jointly optimize multiple objectives, such as accuracy, cost, latency, robustness, and consistency. Ex
vulkan: Tune GDN kernel, fix Intel performance (#29476) vulkan: tune GDN shader tune for Intel Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/51306598 macO
ci: add zdnn backend build but not test (#29541) ci: add zdnn backend build but not test Signed-off-by: Aaron Teo aaron.teo1@ibm.com ci: attempt to run a ubuntu 26.04 container Signed-off-by: Aaron Te
ggml-cuda: HIP: optimize packed byte subtraction (__vsubss4 -> __vsub4) (#29478) ggml-cuda: HIP: optimize non-saturating packed byte subtraction (__vsubss4) CI: ignore 1 spilled vgpr in fattn_vec Co-a
hexagon: optimize concat op (#29673) hex-concat: reduce pkts in gather/transpose hot loop gather directly into dst buffer, use special instruction for gather sync hex-concat: use fastdiv replace calls
model : support classifier_pooling for rerankers (#29627) model : support classifier_pooling for ModernBERT rerankers Assisted-by: Claude Opus 5.5 model : read classifier pooling type in load_hparams
ggml : check row bounds in get_rows_back (#29575) Signed-off-by: Adrien Gallouët angt@huggingface.co Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/5139241
gguf : reject tensor size that wraps after padding (#26979) GGML_PAD(nbytes, alignment) wraps to 0 when nbytes is within (alignment - 1) of SIZE_MAX, which silently bypassed the size overflow guard in
Hexagon f16 activation ops (#29209) hexagon: add F16 support for activation ops (SILU/GELU/GELU_QUICK/GEGLU/SWIGLU) Widens ggml_hexagon_supported_activations() to accept F16 (src0/dst/src1 must agree
ggml-zdnn: impl buffer reset, fix memory leaks (#29637) cont: fix code style Signed-off-by: Aaron Teo aaron.teo1@ibm.com Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/
add GLM-5.3-Flash (GLM5-Next) support (#27773) Rebase GLM-Next support onto master, and migrate to llama-memory-hybrid-idx Add initial MTP support Merge branch optimizations. Reduce allocated compute
musa : define CUDA_ARCH for device passes (#29508) The MUSA vendor header never defined CUDA_ARCH, so every architecture test in the shared ggml-cuda sources evaluated to 0. Kernel bodies gated on the
openvino: serve GET_ROWS on a weight view from the base Constant (#28381) openvino: serve GET_ROWS on a weight view from the base Constant Resolve view_src when collecting weight Constants so a view o
cpu: accept BF16 in src1 of mul_mat (#28937) cpu: accept BF16 in src1 of mul_mat ggml_conv_1d_dw builds its im2col in F32 when the kernel is BF16, then calls ggml_mul_mat(im2col, kernel), which puts F
ggml : add BF16 unary, GLU, binary and scale ops (CPU, CUDA) (#29675) ggml : add BF16 unary, GLU, binary and scale ops (CPU, CUDA) ggml-cpu : use per-op _bf16 functions for BF16 unary and GLU ops Assi
llama: llama_prefetch_rows (#29599) llama: llama_prefetch_rows llama: support row prefetch on Windows Apply the Windows port contributed by @praneshgo unchanged. Source: #29599 (comment) avoid exposin
ci : fix Models Backend Check by shortening the hrm_text fixture (#29744) The fixture recycles its two blocks over 8 cache slots, so the fp16 error builds up past the 1e-4 NMSE bound on the Vulkan T4
jinja : support coerced array attributes (#29574) support coerced array attributes add tests Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/51590476 macOS/
mimo : support dflash (convert + feature extraction) (#29650) convert : update to support dflash cont : fix Co-authored-by: Sigbjørn Skjæret sigbjorn.skjaeret@huggingface.co Co-authored-by: Sigbjørn S
cli: exit on stdin EOF and drop the console wide Ctrl+C broadcast (#29722) cli: exit on stdin EOF and drop the console wide Ctrl+C broadcast On Windows the simple input reader sends CTRL_C_EVENT to ev
ggml/gguf : fix integer overflow (#29384) ggml: fix integer overflow guard for zero-element tensors ggml: validate number of elements in tensor to prevent integer overflow ggml: fix error print Websit
glm5-next: give dead indexer slots unique scatter rows (#29745) The sparse indexer mask is built with a set_rows scatter. Padded pools, absent sequences and missing tail cells all pointed to the same
batch: migrate the rest of examples to llama_batch_ext (#29601) migrate the rest test-thread-safety rm common_batch_staged Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cp
llama: properly handle KV on training (#28520) llama: properly handle KV on training improve Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/51644581 macOS/
test-llama-archs : toggle causal_attn to catch graph shape changes (#29724) After the device decode, flip causal_attn off, decode n_ubatch/2 then n_ubatch tokens. Both have the same node count, so a s
https://preview.redd.it/xs01ds930nsh1.png?width=4800&format=png&auto=webp&s=939762bcf3fbd93e137d4f97e927dda38aaf30fd Back in July I shared LokalBot here. It's a free, open-source Mac app that records
Image-to-3D models redraw your picture, and fine detail - text, logos, faces - comes back garbled or scuffed. My local pipeline that fixes it. (repo link at the end of the post) The latest update adds
Memory chipmaker Micron Technology Inc. predictably delivered another blowout quarter in its latest financial results, with no sign of a slowdown in the demand for artificial intelligence infrastructu
arXiv:2609.35690v1 Announce Type: new Abstract: Robots that learn from a few demonstrations often require two forms of generalization. Compositional generalization recombines skills to solve new tasks
ci : update the oneAPI toolkit to 2026.1 (#29273) ci : update the oneAPI toolkit to 2026.1 oneDNN is removed from Intel Deep Learning Essentials in 2026.0, so staying on the deep-learning-essentials p
common : use fs::path for cache dirs (#29595) Avoid useless string conversions on Windows. No need for BSD or emscripten special cases. Signed-off-by: Adrien Gallouët angt@huggingface.co Website: http
chat : fix Muse Glimmer ignoring response_format json_schema with --jinja (#29615) chat : fix Muse Glimmer ignoring response_format json_schema with --jinja Fixes #29613 chat : accept json fences and
vulkan : reuse descriptor sets when bindings are constant (#29280) vulkan : reuse descriptor sets when bindings are constant vulkan : bump buffer_destroy_count before destroying the buffer Website: ht
ggml : collect all input tensors into graph_inputs (#29634) graph_inputs was populated while splitting the graph, so it only contained the inputs that are used as srcs of some node. With pipeline para
common : add fs_write_atomic() (#29642) Check for buffered write errors when closing downloaded files. Use UTF-8 paths when writing ETag files on Windows. Write in binary mode on Windows. Signed-off-b
common : use fs::path for config dir (#29649) Signed-off-by: Adrien Gallouët angt@huggingface.co Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/51176378 ma
server : remove the built-in UI's service worker when the UI is not served (#29565) With --path or --no-ui, /sw.js returned 404, and a 404 does not remove a service worker, so browsers kept showing th
common : stop accepting draft tokens at EOG (#29638) common : stop accepting draft tokens at EOG cont : remove the test Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/a
hexagon: add FP32 GELU_ERF and GEGLU_ERF support (#29631) hexagon: add FP32 GELU_ERF and GEGLU_ERF support hex-erf: reduce register pressure in kernels Website: https://llama.app Attestations: https:/
ggml : accumulate f16 dot products in f32 on AVX512-FP16 (#29545) Supersedes #29530 Signed-off-by: Adrien Gallouët angt@huggingface.co Website: https://llama.app Attestations: https://github.com/ggml-
vocab : keep NORMAL in PLaMo-2 and PLaMo-3 (#29580) vocab : keep NORMAL in PLaMo-2 and PLaMo-3 The PLaMo-2 and PLaMo-3 vocabularies mark as NORMAL. Current EOG token heuristic matched it by text and a
ggml : fix c++ odr by properly using GGML_COMMON_DECL_CPP (#29504) fix c++ odr by properly using GGML_COMMON_DECL_CPP using actual field rather than macro Co-authored-by: Georgi Gerganov ggerganov@gma
vulkan: MOE aware mat_mul_id tile selection (#29182) mut_mul_id selected its matmul tile with total token count. For MoE dispatch grid the true N per workgroup is per-expert rows. At pp128 on Sarvam 3
vulkan : Load F32 A matrix 2 at a time when its 2-aligned (#29254) It turns out Intel doesn't particularly like loading F32s one at a time and we already have the _2aliagned load logic in mul_mat_vec,
🚨 JUST NOW: Elon Musk and Jensen Huang give an INSIDE LOOK at the big meeting between top AI firms and President Trump today, saying they'll now be 'grading each other's homework' 😆 @elonmusk: 'We sig
this is insane, my bonsai 2 27b build is trending on hugging face, first page of text generation, with 20,964 downloads in its first week. it's prismml's ternary bonsai 2 with the qwen 3.8 mtp head gr
arXiv:2609.32824v1 Announce Type: new Abstract: extit{Gromov-Wasserstein Distances} (GWDs) provide quantitative ways of comparing probabilistic distributions defined on different metric spaces by appl
vulkan: fix wrong results when a mul_mat reads a slice of a larger cache (#28956) vulkan: read the batch stride of an in place src0 from nb[2] A dim01 contiguous tensor can still be a view whose batch
tests : fix ggml init (#29554) tests : init ggml for test-recurrent-state-rollback cont : same for test-save-load-state cont : add to test-state-restore-fragmented + add TODOs Website: https://llama.a
Enables Windows ARM64 build with MSVC cl.exe (#28362) can reproduce the issue vlad sees fix fma issue drop volatile fix volatile runtime task add arm flag if needed fix hsum compile error fix syntax i
context : do not re-reserve the scheduler when toggling causal_attn (#28751) context : do not re-reserve the scheduler when toggling causal_attn llama_context::set_causal_attn() marks the scheduler to
metal: support left and circular padding in GGML_OP_PAD (#29561) metal: support left and circular padding in GGML_OP_PAD Align Metal with CPU, CUDA and Vulkan: shift the source coordinates by the left
ggml-cpu: enable tiled flash attention for non-vector-multiple head dims on x86 (#29423) ggml-cpu: enable tiled flash attention for non-vector-multiple head dims on x86 add AVX2 support for masked loa
tests : refactor test-recurrent-state-rollback (#29426) tests : use llama_context_ptr in test-recurrent-state-rollback Replace raw llama_context pointers with llama_context_ptr and drop the manual lla
webgpu: Handle unaligned writes in ggml_backend_webgpu_buffer_set_tensor (#29471) Fix: Handle unaligned writes in ggml_backend_webgpu_buffer_set_tensor Clang formatting Website: https://llama.app Atte
common : fix HF cache paths on Windows (#29475) Supersedes #29158 Signed-off-by: Adrien Gallouët angt@huggingface.co Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/atte