Knowledge catalogue
Implements the EmbeddingGemma2Model architecture on the MLX runner: 24-layer bidirectional text encoder with PLE, shared gemma4 vision/audio towers, mean-pool + L2 output. /api/embed accepts per-item
llama.cpp build **b8762**, released on April 11, 2026, is centered on the addition of MERaLiON-2 multimodal audio support to the project's `mtmd` (multimodal) framework. It adds support for A*STAR's M
server : support typed content (vision/audio/video) input for /v1/embeddings endpoint (#29556) server : support multimodal input for /v1/embeddings (Qwen3-VL-Embedding) Accept the OpenAI-style wrapped
Overview llama.cpp 0.3.0 introduces the dots3-note multimodal model (with a new DSA-ISWA KV cache), MTP support for GLM-4.5-Air, and tensor-split (-sm tensor) plus multi-sequence rollback fixes for De
Release b9503 of llama.cpp fixes multimodal (mtmd) support by handling Gemma 4 audio projector embedding size , specifically removing the projection_dim from clip_n_mmproj_embd. This build includes pr
Release b9169 of llama.cpp includes updates to multi-token multimodal decoding (mtmd) functionality, adding chunks and fixing preprocessing for Qwen3A models . The changes include attention mask imple
Release b9122 of llama.cpp addresses precision issues for multimodal models through ggml-webgpu, including fixes for GELU functions, flash attention tile implementations, and type conflict resolution.
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
server : allow RANK pooling batch splitting for causal LLM rerankers (ie. Qwen3 and Qwen3-VL) (#28876) server : allow splitting RANK pooling for causal LLM rerankers Rerank models fall into two catego
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.
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
Muse Glimmer Note: Muse Glimmer is currently available via initial support via Ollama's MLX engine on Apple Silicon. Additional support and optimizations for Apple Silicon, NVIDIA, AMD, and other plat
This release candidate introduces support for offloading the image GPU projection (mmproj) to an integrated GPU when using fit padding in Ollama's LLM processing, addressing technical improvements for
Build **b8772** is a recent incremental release of [llama.cpp](https://github.com/ggml-org/llama.cpp), the open-source C/C++ library for local LLM inference. Based on surrounding release activity, it
Build **b8779** is an incremental release of [llama.cpp](https://github.com/ggml-org/llama.cpp), an open-source C/C++ framework for running large language model inference locally and in the cloud. Lik