Search: “model-releases”

The decode loop releases MLX's pool of freed buffers every 256 generated tokens, which is also how often the KV cache grows and drops its previous, smaller buffers. The check fires only when the token

llama-bench: support --version to print build info (#28971) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/47831792 macOS/iOS: macOS Apple Silicon (arm64)

ci : add self-hosted webgpu to hf-jobs (#28712) add self-hosted vulkan and webgpu to hf-jobs try t4-medium cont : adjust cpu backend threads try t4-small again restore cm jobs Co-authored-by: Georgi G

metal: fix NaN in mul_mm_id when activations exceed f16 range (#26223) test-backend-ops: reproduce MUL_MAT_ID NaN for activations beyond f16 The Metal mul_mm_id path narrows src1 to half for the simdg

vulkan: make MUL_MAT_ID BN/2 tail unconditional (#28923) Use BN/2 as the default for BNover2 and as the disabled fallback for BNover4, and remove the enable gate from the MUL_MAT_ID BN/2 branch. The B

chat : force n</think> on reasoning budget end for qwen3-coder (#28869) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/47858837 macOS/iOS: macOS Apple Sili

HIP: broaden MoE ncols_opt tile heuristic on RDNA3.5 architecture (#28935) It's found the MoE ncols_opt tile heuristic needs to be broadened to include the RDNA3.5 architecture. The code change is imp

qwen4exp: add hc ops (#28901) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/47897353 macOS/iOS: macOS Apple Silicon (arm64) macOS Apple Silicon (arm64, Kl

Change max context length for auto-fitting with unified KV (#28849) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/47906437 macOS/iOS: macOS Apple Silicon

rpc : invalidate cached compute graph when a referenced buffer is freed (#24292) The server caches the most recent compute graph per device so that GRAPH_RECOMPUTE can re-execute it without resending

spacemit : fix wrong transpose function for int16 data (#25161) The sizeof(int16_t) branch in permute_transpose_impl calls rvv_transposed_s32_mn_to_nm instead of rvv_transposed_s16_mn_to_nm. This is a

CUDA/HIP: improve access patterns in im2col (#28013) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/47937267 macOS/iOS: macOS Apple Silicon (arm64) macOS A

model : add support for HrmTextForCausalLM (DFM Mimir 1B) (#27625) model : add support for HrmTextForCausalLM (DFM Mimir 1B) HRM-Text runs two transformer stacks (low, high) in an alternating cycle ov

hexagon: accept the zeroed rope probe in supports_op (#28995) llama probes weight placement with a rope where all params are 0, so rejecting n_dims == 0 or freq_base == 0 puts rope_freqs on the CPU. T

hexagon: Support for K-Quants Q4_K and Q6_K (#28994) implement q6k/q4k kernels Squashed from: feat: implement q6k kernel hex-q6k: improve unpack accuracy hex-q4_k: add support for Q4_K kernels Co-auth

Enable CUDA graph for MTP draft (#28549) Improve CUDA graph usage for MTP Rename field Address review feedback Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestatio

TP: fix split state and granularity for fused QKV gemma4, qwen35 (#28965) model: calculate split states for attn_qkv from n_head * n_embd_head_k required for gemma4 with --fuse-qkv, where n_embd is 53

model is one package with three jobs: the contract between the runner and the architectures, the opened checkpoint, and building nn layers from checkpoint tensors. Its files did not say which was whic

ci : fix android release (#28936) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/47548995 macOS/iOS: macOS Apple Silicon (arm64) macOS Apple Silicon (arm64

ci: Bump CUDA Windows x64 builds to 13.4.1 (#28930) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/47583571 macOS/iOS: macOS Apple Silicon (arm64) macOS Ap

metal : add FA kernels for HSK=96, HSV=64 (MiniCPM3) (#28599) metal : add FA kernels for HSK=96, HSV=64 (MiniCPM3) MiniCPM3 sets attention.key_length to 96 and does not set attention.value_length, whi

opencl: add generic ssm_scan (#28881) opencl: add generic ssm_scan opencl: fix whitespace Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/47601188 macOS/iOS

OpenVINO: optimize stateful decode and GPU MoE inference (#28638) exclude GPU/NPU failing POOL_2D case Fix pool case ggml-openvino: fix stateful decode for Gemma-4 per-layer-type head sizes ggml-openv

vulkan: support sparse Flash Attention (#28105) vulkan: add sparse Flash Attention support for DSV4/GLM tune implementation add tests avoid nondeterministic atomicAdd add cm2 decode vector support sim

models : move build_arch_graph() after graph() template specialization (#28934) Move build_arch_graph()'s function definitions after the graph and graph template specializations have been explicitly d

cuda: support row-contiguous SUM_ROWS (#26308) cuda: support row-contiguous SUM_ROWS organize the code and add GGML_OP_MEAN to support row-contiguous tensors using the same shared kernel, and add a te

rpc : hash-cache only weights (#28789) rpc : hash-cache only weights ggml_backend_rpc_buffer_set_tensor and ggml_backend_rpc_set_tensor_async hashed every transfer above HASH_THRESHOLD and let rpc-ser

opencl: choose the MoE expert matmul by batch size for speculative decoding/MTP (#27637) opencl: gate the prebuilt q4_0 MoE GEMM on routing count opencl: stop writing zeros into the padded MoE activat

HIP: Enable AllReduce for ROCm (#27825) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/47762387 macOS/iOS: macOS Apple Silicon (arm64) macOS Apple Silicon

hex-cpy: use dma if src and dst are contiguous (#28906) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/47770889 macOS/iOS: macOS Apple Silicon (arm64) macO

hexagon: add back missing contiguous fast-path and hvx_copy_uu for each run (#28886) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/47775478 macOS/iOS: mac

What's Changed app: fix ChatGPT model selector spacing mlxrunner: Evict prefix cache snapshots from the active conversation mlxrunner: check system free memory and wait for evicted runners before load

**Release Tag:** v0.34.2‑rc0 (commit `b10969`) **Key Details:** This release candidate bumps the version of *llama.cpp* and addresses a build issue where duplicate symbols were generated between `libl

ggml-cuda: fallback to F32 on device without BF16 hardware acceleration (#28846) ggml-cuda: fallback to F32 on device without BF16 hardware acceleration: (Nvidia >= AMPERE, AMD >= RDNA3 or = CDNA) app

common : move llama_n_rs_seq to before llama_decode (#28749) This commit moves the llama_n_rs_seq function call to before the llama_decode call and returns directly if the check is true, removing the

sycl : fix oneDNN scratchpad breaking the pool free order (#28704) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/47287268 macOS/iOS: macOS Apple Silicon (

ggml-cpu : disable PCH and fix CACHE_LINE_SIZE ambiguity to fix heap corruption (#28882) Disable the ggml-cpu precompiled header and remove the std::hardware_destructive_interference_size branch from

sycl: rfc: Use radix select for top_k (#28670) sycl: GPU-resident TOP_K for large k, parallelised over the device The SYCL backend refused GGML_OP_TOP_K above k = 32 and let it fall back to the CPU, a

llama.cpp : bump version to 0.4.1 (#28900) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/47376843 macOS/iOS: macOS Apple Silicon (arm64) macOS Apple Silic

ci : add ubuntu-cuda builds to release (#28186) release : add ubuntu-cuda build job (12.8/13.3, x64+arm64) Add GCC 14 for CUDA arm64 builds in CI Eplicit bash Install git for CCCL fetch Install git be

HIP: fattn-mma: use fp32 accumulation on MFMA devices (#28576) use fp32 accumulators in fattn-mma on CDNA Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/47

Overview llama.cpp 0.4.1 adds Maple 20B-A1B, Tencent Hy 4, and Spark2.5 support, improves JSON schema handling, chat parsing, logging, and server child-process management, and updates ggml to v0.24.0.

common : implement common_schema internal representation for JSON schemas (#28736) common : implement common_schema types common : implement a json schema optimizer common : reduce optimizations commo

common: add LOG_JSON macro to log structured data (#28586) add LOG_JSON macro fit: add demo LOG_JSON Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/4711287

chat : improve parsing of complex types in qwen3-coder (#28742) chat : improve schema support in qwen3 parser cont : clean up grammar a bit Website: https://llama.app Attestations: https://github.com/

opencl: apply the noshuffle row-alignment rule to q4_K, q5_K and q8_0, not just q6_K (#28575) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/47132459 macOS