AI NEWS JOURNAL

AI, IN PERSPECTIVE

Search AI news

Search headlines, summaries and articles to connect the dots.

2026-11-08

llama.cpp releases05:11

b12191

The latest update to llama.cpp introduces enhanced MoE (Mixture of Experts) cache functionality, enabling computation of cache misses on the CPU with a configurable CPU share. A new tuning script, moe-cache-tune.py, automatically determines optimal cache size and CPU share settings based on system performance. The update also improves scheduling efficiency, removes redundant sync operations, and refactors code for better maintainability, while maintaining compatibility across multiple platforms including macOS, Linux, Windows, and Android.

llama.cpp releases16:11

b12202

The llama.cpp v1.220.2 release introduces enhanced BF16 precision support across multiple components, aligning with reference implementations by moving specific operations like GLM4 family projections, MiMoVL vision matmuls, and Qwen3-TTS speaker encoder conv1d to BF16 where appropriate. Backends such as CUDA, Hexagon, OpenCL, WebGPU, and Vulkan now uniformly handle GGML_PREC_BF16 similarly to F32, while CPU, SpaceMIT, and ET backends extend flash attention support to BF16 with F32 accumulation. Flash attention tests now include BF16 cases, and the KQ multiplication in Llama is updated to request BF16 precision. A TODO remains in clip.cpp for future work.

llama.cpp releases22:20

b12207

A new fast path for d_state=64 has been added to the SSM_SCAN kernel in ggml-cuda, improving generation throughput for Zamba2-7B-Instruct from 11 to 15.1 tok/s—approximately a 35% increase—by eliminating CPU fallbacks for this state size. The change was verified with real-world Zamba2 hyperparameters and maintains correctness, though a separate 28% of scan calls still fall back to CPU due to an undiagnosed issue.

llama.cpp releases23:47

b12208

The ggml library has been updated to make tied selections in the radix Top-K algorithm deterministic on CUDA and Metal backends by replacing atomic compaction loops with a deterministic scheme using simdgroup scans and coalesced memory accesses. This change ensures byte-identical outputs across runs, improves performance by 20-45% on M5 Ultra hardware, and enables deterministic testing for Metal and ROCm. Redundant CUDA kernel resets were removed, and Vulkan/Metal changes were reduced or skipped.

2026-11-07

llama.cpp releases00:39

b12176

The update qualifies calls to math functions like isnan and signbit with std:: to ensure compatibility with C++ standard library overloads, addressing issues on NetBSD 11.99.9/x86_64 with gcc 14.4.0. Additional C math function calls are standardized using std:: prefixes (e.g., std::fabs, std::copysign) instead of non-existent std::fabsf variants, improving accuracy by keeping computations in double precision. The change results in a minor behavioral difference with a relative error of at most 1.4e-7, well below test tolerances.

llama.cpp releases19:58

b12188

The server update addresses slot processing order by ensuring parent slots are visited before child slots, preventing delays caused by child tasks landing on lower slot IDs. This change ensures all child tasks join the same batch after being woken by their parent. Additionally, new test tags have been introduced for monitoring: __TEST_TAG_WAIT_PARENT_SLOT__ for child slot waits, __TEST_TAG_PROMPT_CACHE_RESTORE__ for prompt cache restores, and __TEST_TAG_PROMPT_CACHE_FULL_RESTORE__ when a cached prompt is fully contained in a new one based on token counts.

llama.cpp releases20:15

b12189

The metal backend's kernel_group_norm_f32 was updated to correctly handle tensors with ne03 > 1 and non-contiguous inputs by dispatching one threadgroup per (group, i03) pair, computing per-batch statistics, clamping group element count for incomplete groups, and supporting strided access via separate strides for source and destination. The fix also restricts GROUP_NORM to f32 inputs and improves performance through specialized pipeline variants.

2026-11-06

llama.cpp releases10:48

b12143

The Vulkan implementation of the Fast Walsh-Hadamard Transform (FWHT) has been enhanced to support block widths up to 8192, using shared-memory shaders for larger blocks, and now includes F16 input capability via shader-float16, with pipelines created only when device limits allow. This update aligns Vulkan with CUDA's capabilities and improves efficiency by eliminating the need for F32-to-F16 conversions.

2026-11-04

llama.cpp releases10:40

b12094

Optimized TOP_K performance on Hexagon backend by skipping bitonic sort for tiles whose maximum cannot surpass the k-th best value, replacing tile sort with top_k_scan() to reduce 256-float blocks using HVX max-reduce, and inserting only surviving elements into a sorted k-entry list. Supports k values from 2 to 64 and row widths over 1024 in both single-row and multi-row paths. NaN treated as -INF, ties preserve first-seen index. Eliminated scalar reads from VTCM in hex-top-k.

llama.cpp releases15:59

b12100

The update resolves issues with -hf tag matching in longer quantization names, ensuring tags start at token boundaries. It also adds support for underscores as separators in quantization tags, improving compatibility with filenames like step3p5_flash_Q4_K_S-00001-of-00012.gguf.

llama.cpp releases17:12

b12102

The ggml library now checks view bounds based on the actual bytes reached by strides, improving accuracy for strided views. Views are required to start from contiguous elements along dimension 0, ensuring correct data access. Convolution windows in delta_net are created with all rows at once to prevent stride-related issues.

2026-10-28

llama.cpp releases09:58

b11946

server : add realtime conversation and transcription on /v1/realtime (#31269) * server : add realtime conversation and transcription on /v1/realtime Co-authored-by: Xuan Son Nguyen Assisted-by: Claude Opus 5.5 * examples : default the realtime client to the server port 9931 Assisted-by: Claude Opus 5.5 * server : sample realtime responses right after their own decode Slots respond in turns on one context: sampling at the start of the next step read the logits of whichever slot decoded last, so concurrent sessions mixed their outputs. Assisted-by: Claude Opus 5.5 * move more stuff to server-realtime, making it self-contained * examples : fix the type check of the realtime client Assisted-by: Claude Opus 5.5 * examples : use the model card prompts in the realtime client Assisted-by: Claude O

Timeline and related coverage

2026-10-27

llama.cpp releases20:30

b11928

model: Add MiniCPM-OCR-Jev (#30980) * model : add MiniCPM-OCR-Jev support The model is a Qwen3.5-VL trunk with a value head read at the last token: it scores an OCR text against an image rendered from it, it does not generate. Add the openbmb-minicpm-ocr-jev decision type, served by POST /v1/systemone. The class scores are the pooled output of the sequence, read with last pooling, so the conversion turns score.weight into cls.output and sets embedding_length_out. The checkpoint has no LM head, so text generation is rejected instead of using the output weights duplicated from the token embeddings. The task is fixed, so its prompt is entirely in the systemone template, the same way h2o-lightning keeps its default there. The processor of the checkpoint allows up to 16384 merged tokens per ima

Timeline and related coverage

llama.cpp releases16:50

b11923

ggml : relax scale to non-contiguous views (#31146) * ggml : relax scale to non-contiguous views SCALE requires src0 to be 1D-contiguous. Relax this to dim-0 contiguity with arbitrary row strides (ggml_is_contiguous_rows), same as the unary ops in the Metal backend. - ggml: relax the assert in ggml_scale_impl - cpu: address src0 rows through their byte strides (f32, bf16) - cuda: stride-aware kernel, fastdiv row decomposition - metal: support bf16 scale (guarded unary instantiations) - sycl: keep the gate at contiguous, the kernel is flat f32 Add view variants to test_scale. Verified on CPU, Metal and CUDA. Assisted-by: dsh:llama.cpp/ggml-org/Qwen3.8-Flash-Next-GGUF (RTX 5090 32GB, DDR5 128GB) * cuda : skip the softcap fusion on non-contiguous views The fused softcap kernel reads src and w

Timeline and related coverage

llama.cpp releases16:18

b11922

tests : fix multi-seq split replay and partial-state cases for trained models (#31216) * tests : fix multi-seq split replay against trained models The rollback snapshot is captured by the graph that decoded the rolled-back tail, while the reference context decodes the same prefix with different batch shapes. The two states differ in the last bits and trained models amplify the difference with every replayed token. Compare only the first replayed token of each seq, then restore the rollback state into the reference context before the seq-1-only tail. Assisted-by: dsh:llama.cpp/ggml-org/Qwen3.8-Flash-Next-GGUF (MacBook, M5 Max, 128GB) * tests : require bit-exact seq-1 isolation in split replay The seq state save is not always bit-lossless - SWA-masked cells are not saved - so reload the rest

Timeline and related coverage

2026-10-26

llama.cpp releases07:28

b11888

common, grammar: add recursive depth limit, make it a requirement (#31153) * common, grammar: add recursive depth limit, make it a requirement * catch more recursive functions * common : tidy up peg parser changes * grammar : reject candidates in a loop, without a depth limit The depth limit in llama_grammar_reject_candidates_for_stack counted the code points of a vocab token, so any token of 256 chars or more was always rejected under a grammar, such as the 512-space token of GLM-4.7-Flash or the long space tokens of Laguna-S-2.1. The rejection now walks the code points with its own stack of levels instead of recursing: there is no limit and no stack risk, whatever the length of a token from the GGUF, and the masks match master. * grammar : track rejected candidates as indices into the in

Timeline and related coverage

llama.cpp releases00:12

b11879

CUDA: fuse FFN gate/up matmuls and GLU in MMQ with interleaved gate/up rows (#29948) * cuda: fuse FFN gate/up and GLU in MMQ with interleaved gate/up rows For prompt processing, run the gate and up matmuls of a dense FFN as one mul_mat_q launch and apply the GLU in the write-back. - The loaders take the gate weights as a second pointer and fill the weight tile with up and gate rows interleaved in chunks of 8, so each thread holds an up value and the matching gate value (rows r and r + 8 of an MMA tile). - The MMA loop is unchanged: same tile, warp layout, vec_dot and one sum[] per thread. The fused write-back computes the GLU from registers and writes half the rows. - The fused path uses the regular tiling (no stream-k) with the column tiles innermost so that neighbouring blocks share weig

Timeline and related coverage