Script 'mail_helper' called by obssrc Hello community, here is the log from the commit of package llamacpp for openSUSE:Factory checked in at 2026-10-01 22:20:17 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Comparing /work/SRC/openSUSE:Factory/llamacpp (Old) and /work/SRC/openSUSE:Factory/.llamacpp.new.1253 (New) ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Package is "llamacpp" Thu Oct 1 22:20:17 2026 rev:36 rq:1381923 version:0.5.0 Changes: -------- --- /work/SRC/openSUSE:Factory/llamacpp/llamacpp.changes 2026-09-18 22:09:02.811900862 +0200 +++ /work/SRC/openSUSE:Factory/.llamacpp.new.1253/llamacpp.changes 2026-10-01 22:21:07.194631172 +0200 @@ -1,0 +2,18 @@ +Thu Sep 24 14:17:52 UTC 2026 - Robert Munteanu <[email protected]> + +- Update to version 0.5.0 + * Add HRM-Text, MiMo-V2.6, and HunyuanOCR support; extend Nemotron + MTP and Nemotron-H handling. + * Improve CUDA, Metal, Vulkan, SYCL, OpenCL, OpenVINO, Hexagon, + and WebGPU backends, including CUDA conv2d and Metal + MoE/SSM_CONV fusion. + * Add multi-address server binding, image function-call outputs, + and improve router lifecycle, token counting, and startup logging. + * Improve chat parsers, JSON schema handling, multimodal validation, + and model save/load support. + * Update the bundled GGML library to 0.25.1 with sparse flash + attention, IQ4_XS Vulkan kernels, and Metal f32 x bf16 matmul variants. + * Include the b11146 Web UI asset with WEBM support and responsive + layout, toast, and reasoning-menu fixes. + +------------------------------------------------------------------- Old: ---- llama-b10964-ui.tar.gz llamacpp-0.4.1.tar.gz New: ---- llama-b11146-ui.tar.gz llamacpp-0.5.0.tar.gz ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Other differences: ------------------ ++++++ llamacpp.spec ++++++ --- /var/tmp/diff_new_pack.lDuPkD/_old 2026-10-01 22:21:08.634692154 +0200 +++ /var/tmp/diff_new_pack.lDuPkD/_new 2026-10-01 22:21:08.636692238 +0200 @@ -18,7 +18,7 @@ %global backend_dir %{_libdir}/ggml -%global upstream_build 10964 +%global upstream_build 11146 %global license_dir %{_datadir}/licenses %global llama_sover %{version} @@ -27,7 +27,7 @@ %global mtmd_sover %{llama_sover} %global mtmd_sover_suffix 0 -%global ggml_sover 0.24.0 +%global ggml_sover 0.25.1 %global ggml_sover_suffix 0 %if 0%{?suse_version} == 1500 @@ -43,7 +43,7 @@ %endif Name: llamacpp -Version: 0.4.1 +Version: 0.5.0 Release: 0 Summary: Inference of Meta's LLaMA model (and others) in pure C/C++ License: Apache-2.0 AND MIT AND BSD-2-Clause AND BSD-3-Clause AND ISC AND MPL-2.0 ++++++ llama-b10964-ui.tar.gz -> llama-b11146-ui.tar.gz ++++++ ++++ 7613 lines of diff (skipped) ++++++ llamacpp-0.4.1.tar.gz -> llamacpp-0.5.0.tar.gz ++++++ /work/SRC/openSUSE:Factory/llamacpp/llamacpp-0.4.1.tar.gz /work/SRC/openSUSE:Factory/.llamacpp.new.1253/llamacpp-0.5.0.tar.gz differ: char 22, line 1
