Download Latest Version llama-b9484-bin-ubuntu-openvino-2026.0-x64.tar.gz (12.9 MB)
Email in envelope

Get an email when there's a new version of llama.cpp

Home / b9481
Name Modified Size InfoDownloads / Week
Parent folder
llama-b9481-xcframework.zip < 11 hours ago 200.0 MB
llama-b9481-ui.tar.gz < 11 hours ago 1.9 MB
llama-b9481-bin-win-vulkan-x64.zip < 11 hours ago 33.1 MB
llama-b9481-bin-win-opencl-adreno-arm64.zip < 11 hours ago 10.3 MB
llama-b9481-bin-win-hip-radeon-x64.zip < 11 hours ago 319.7 MB
llama-b9481-bin-win-cuda-13.3-x64.zip < 11 hours ago 158.4 MB
llama-b9481-bin-win-cuda-12.4-x64.zip < 11 hours ago 260.3 MB
llama-b9481-bin-win-cpu-x64.zip < 11 hours ago 16.1 MB
llama-b9481-bin-win-cpu-arm64.zip < 11 hours ago 9.7 MB
llama-b9481-bin-ubuntu-x64.tar.gz < 11 hours ago 14.5 MB
llama-b9481-bin-ubuntu-vulkan-x64.tar.gz < 11 hours ago 32.3 MB
llama-b9481-bin-ubuntu-vulkan-arm64.tar.gz < 11 hours ago 25.5 MB
llama-b9481-bin-ubuntu-s390x.tar.gz < 11 hours ago 13.0 MB
llama-b9481-bin-ubuntu-rocm-7.2-x64.tar.gz < 11 hours ago 130.0 MB
llama-b9481-bin-ubuntu-openvino-2026.0-x64.tar.gz < 11 hours ago 12.9 MB
llama-b9481-bin-ubuntu-arm64.tar.gz < 11 hours ago 11.6 MB
llama-b9481-bin-macos-x64.tar.gz < 11 hours ago 9.9 MB
llama-b9481-bin-macos-arm64.tar.gz < 11 hours ago 9.7 MB
llama-b9481-bin-android-arm64.tar.gz < 11 hours ago 73.8 MB
cudart-llama-bin-win-cuda-13.3-x64.zip < 11 hours ago 391.0 MB
cudart-llama-bin-win-cuda-12.4-x64.zip < 11 hours ago 391.4 MB
b9481 source code.tar.gz < 12 hours ago 34.0 MB
b9481 source code.zip < 12 hours ago 35.5 MB
README.md < 12 hours ago 4.4 kB
Totals: 24 Items   2.2 GB 0
model : support granite multilingual embeddings R2 (ibm-granite/granite-embedding-{97,311}m-multilingual-r2) (#22716) * Add support for the ibm-granite/granite-embedding-{97m,311m}-multilingual-r2 embedding models: * Added a version of the gpt4o tokenizer that has a fixed regex (better handling of marks), and different token merging setting for the 97m model * Reused gemma4 tokenizer for the 311m model * granite-embedding-*-multilingual-r2 : add support SwiGLU FFN for Granite Embedding Multilingual R2 * added new GGUF key <arch>.hidden_activation (LLM_KV_HIDDEN_ACT) + writer * added a forward declaration of llm_ffn_op_type to llama-hparams.h * added llm_ffn_op in hparams * added LLM_FFN_NONE = 0 sentinel to llm_ffn_op_type (value-initialization), modern-bert: explicitly assigns LLM_FFN_GEGLU before reading GGUF (unchanged). * centralized hidden_act mapping in llama-model.cpp, added llm_ffn_op_type_from_string() helper, mirroring rope_scaling_type/llama_rope_scaling_type_from_string() * modern-bert reads the GGUF key (when present) and uses the resulting op in its FFN graph * Added granite-embedding-{97m,311m}-multilingual-r2 to the converter code * Added the hashes for the granite embedding multilingual R2 models * Set the hidden_activation in the GGUF if the field is present in config.json (such as for the granite embedding models)

macOS/iOS:

Linux:

Android:

Windows:

openEuler:

  • DISABLED
  • openEuler x86 (310p)
  • openEuler x86 (910b, ACL Graph)
  • openEuler aarch64 (310p)
  • openEuler aarch64 (910b, ACL Graph)

UI:

Source: README.md, updated 2026-06-02