b10660
model: add Qwen3.8-Flash-Next (qwen4exp) (#27742) gguf: add qwen4exp (Qwen3.8-Flash-Next) arch and converter Adds the GGUF-side plumbing for HF model_type qwen4_exp: MODEL_ARCH.QWEN4EXP plus tensors f
Knowledge catalogue
model: add Qwen3.8-Flash-Next (qwen4exp) (#27742) gguf: add qwen4exp (Qwen3.8-Flash-Next) arch and converter Adds the GGUF-side plumbing for HF model_type qwen4_exp: MODEL_ARCH.QWEN4EXP plus tensors f
fit: also take into account n_streams (#27496) fit: also take into account n_streams server: make the draft context follow the target context With a non-unified KV cache the target context now holds n
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
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
metal : single-source fusion table + fusion debug rework (#28164) metal : rework fusion patterns into a single table All fusable op patterns for the Metal backend are now declared once in a fusion tab
model : fix MTP context kv cache allocation for deepseek2, glm4moe, … (#28630) model : fix MTP context kv cache allocation for deepseek2, glm4moe, cohere2moe architectures (#28626) model: add inverse
Overview llama.cpp 0.4.0 adds initial Qwen3.8-Flash-Next and Nemotron-3-Puzzle support, on-demand tensor reading, per-slot server context limits, video input options, and a ggml update to 0.23.0 with
mtmd : add const in various places (#28307) mtmd : mark context as const in more methods Mark mtmd_context as const in: mtmd_bitmap_init_lazy mtmd_tokenize mtmd_tokenize_from_parts mtmd_helper_support
sycl: make --fit respect --fit-target better (#27629) improve the --fit algorithm to take into account the actual peak required VRAM for a given context size on a SYCL backend. This includes both prop
tests : run test-save-load-state across all architectures (#27755) tests : run test-save-load-state across all architectures test-save-load-state previously only ran in ctest against a single download
sampler : remove 'full-context windows' from history-based samplers (#26524) Resolve -1 to 1024 instead of ctx-len for samplers Because of backend-sampling we initialize samplers before the complete l
Release b9292 of llama.cpp fixes a memory leak in the server context where speculative decoder, draft context, and draft model were not properly freed during destroy(), causing VRAM leaks on sleep/res
ggml-openvino: update to 2026.4.1, optimize performance, expand ops, improve device listing. (#29852) ggml-openvino : Qwen3.5 MoE perf (#312) Squash of ravi9#312: ggml-openvino: add detailed inference
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
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
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
feat: add GLM5Next MTP, optimize (#29928) llama : add GLM5-Next NextN (MTP) graph Build the GLM5-Next multi-token-prediction head as graph_mtp: the NextN block embeds enorm(tok)+hnorm(h) through eh_pr
llama : fix unexpected graph reallocation in the k-pool models (#29958) llama : fix unexpected graph reallocation in the k-pool models Both k-pool models built a graph shape that depends on state the
Overview llama.cpp v0.6.0 introduces the new llama_batch_ext extended batch API (with llama_process) for mixed token/embedding inputs and MTP/deepstack state embeddings, adds support for the GLM-5.3-F
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.
CUDA: Add backend sampler for penalties sampler (#25262) sampling: enhance penalty handling in common_sampler_init Set default value for penalty_last_n based on model context if not specified. Ensure
model: add NextN/MTP speculative decoding support for GLM_DSA (GLM-5.2) (#25980) model: add NextN/MTP speculative decoding support for GLM_DSA (GLM-5.2) Adds GLM-5.2 NextN/MTP as a --spec-type draft-m
llama : re-reserve the sched when the nextn extraction flags change (#30020) The speculative MTP init enables NextN extraction on the target and draft contexts after both were created and their schedu
Revert 'Change max context length for auto-fitting with unified KV (#28849)' (#29437) This reverts commit b04d4e5. Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attest
Overview This release focuses on backend performance and correctness, broader model coverage, and more robust server/router operation. It adds HRM-Text (DFM Mimir 1B) support, MiMo-V2.6 and HunyuanOCR
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
vulkan: use CPU writes in ggml_backend_vk_cpy_tensor_async if the context is idle (#28618) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/46709648 macOS/iO
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
memory : copy Hadamard matrix to k_rot tensor only if it has buffer assigned to prevent crashes during context shift of unquantized K cache (#27967) Co-authored-by: Stanisław Szymczyk sszymczy@gmail.c
Learn how to write your first prompt in the GitHub Copilot app, choose the right context and model, and start your first task with confidence. The post Write your first prompt with the GitHub Copilot
llama : support multi-output backend sampling (#25532) Enable backend sampling with token speculation Clamp the mask sum before converting it into the sampled index Add a numeric context parameter dec
opencl: bugfix increment ref_count in ggml_backend_opencl_init() (#26162) Incrementing ref_count at the beginning is important later in the free() method of the ggml_backend_opencl_context at program
ggml-webgpu: improve flash_attn_vec for quantized KV at long contexts (#25956) improve fa of quantized kv cache Fix some bugs and some comments. fix v type check and some comments Fix build error caus
llama-context : sync pending async copies before clearing embd_seq (#25676) Website: https://llama.app macOS/iOS: macOS Apple Silicon (arm64) macOS Apple Silicon (arm64, KleidiAI enabled) DISABLED mac
The search did not return the specific details of the b9947 release. Based on the repository and context available: B9947 is a release of llama.cpp, which enables LLM inference in C/C++ . The release
The search results do not contain specific information about v0.32.0-rc0. Based on the version numbering and context from Ollama's release patterns, v0.32.0-rc0 is a release candidate for Ollama, a la
The search results do not contain specific details about release b9938. However, based on the context available, b9938 is one of the regular build releases from the llama.cpp project. Llama.cpp releas
The search results don't contain specific details about the b9914 release. Based on the context of llama.cpp releases, b9914 is a build/commit version in the llama.cpp project, an open-source tool for
The search results don't provide specific details about release b9929. Based on the available information and the context that llama.cpp releases frequently with tagged versions, b9929 is a specific b
The search results don't contain specific details about release b9871. Based on the pattern evident in the search results and the context, here's the summary: B9871 is a release of llama.cpp, an open-
The search results show llama.cpp releases but do not contain specific information about release b9853. Based on the context from the llama.cpp project, b9853 is likely a development build release of
b9775 is a release of llama.cpp published on June 23, 2026 , featuring 'server: check draft context creation error' improvements . The release includes pre-built binaries across multiple platforms inc
The search results don't contain specific information about release b9584. Based on the context from other llama.cpp releases in the results, b9584 is likely an intermediate build version of llama.cpp
The search results did not provide specific information about the b9564 release. Based on the context of llama.cpp releases and the pattern observed with nearby releases (b9542, b9543, b9544, etc.), I
The search did not return specific details about the b9547 release. Based on the repository context and release naming convention, b9547 is a build number for llama.cpp, an open-source C/C++ implement
The search results show general llama.cpp information and references to other recent builds (like b9484), but the specific details for b9487 were not clearly accessible. Based on the context from llam
The search results do not contain specific details about the b9494 release. Based on the context from the llama.cpp GitHub releases page, b9494 is an intermediate build release of llama.cpp, a C/C++ p
The search results do not contain specific information about release b9496. Based on the available context, b9496 is a build/release version of llama.cpp, a C/C++ implementation for efficient LLM infe
The search results show recent llama.cpp releases but do not contain specific information about release b9467. Based on the context of llama.cpp releases and the GitHub repository structure, b9467 is
The search results don't contain specific information about release b9470. Based on available context, b9470 is a build release from llama.cpp, a C/C++ implementation that enables LLM inference with m
The search results do not contain specific details about release b9354. Based on the context from the llama.cpp project, b9354 is a build release of llama.cpp, a C/C++ implementation for running large
The search didn't provide specific details about release b9365. Based on the GitHub repository pattern and context, b9365 is a build release of llama.cpp, which is a C/C++ implementation for enabling
The search results show information about llama.cpp releases but don't contain specific details about build b9297. Based on the context and numbering (b9297 is a recent build after b9294), b9297 is li
Release b9260 of llama.cpp includes OpenCL backend refactoring that improves initialization, GPU identification, and performance by caching global memory size in device context. llama.cpp enables LLM
I was unable to find specific details about the b9213 release in the search results. Based on the context from the llama.cpp project, b9213 is an intermediate build release of llama.cpp, the C/C++ imp
The search did not return specific information about the b9140 release. Based on the context, b9140 is a specific build number from the llama.cpp project releases. llama.cpp is an LLM inference implem
The search results don't contain specific information about release b9087. Based on the context, b9087 is a release from llama.cpp, a C/C++ implementation for LLM inference . This release likely conta
I was unable to find specific details about the b9058 release. Based on the context, b9058 is a build version from the llama.cpp project, which is an open source software library that performs inferen
The search did not return specific details about release b8982. Based on the llama.cpp project context from the search results, b8982 is a build/release version from the llama.cpp repository that foll
The search results do not contain specific information about release b8948 itself. Based on the context from llama.cpp releases, b8948 is one of the continuously updated build releases from the llama.