Script 'mail_helper' called by obssrc
Hello community,

here is the log from the commit of package ollama for openSUSE:Factory checked 
in at 2026-10-23 20:29:13
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Comparing /work/SRC/openSUSE:Factory/ollama (Old)
 and      /work/SRC/openSUSE:Factory/.ollama.new.2252680 (New)
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Package is "ollama"

Fri Oct 23 20:29:13 2026 rev:72 rq:1387265 version:0.40.9

Changes:
--------
--- /work/SRC/openSUSE:Factory/ollama/ollama.changes    2026-10-20 
12:40:57.639114643 +0200
+++ /work/SRC/openSUSE:Factory/.ollama.new.2252680/ollama.changes       
2026-10-23 20:30:19.452679693 +0200
@@ -1,0 +2,50 @@
+Fri Oct 23 07:38:51 UTC 2026 - Adrian Schröter <[email protected]>
+
+- Updated to version 0.40.9
+  * Add speech generation with Phonon on Apple Silicon
+  * ollama run plays audio as text or reply is generated
+  * OpenAI /v1/audio/speech can stream audio
+  * Add d3 decision models from vLLM Semantic Router
+  * d3 classifies, answers yes/no, or scores options
+  * d3 comes in 2B, 4B, 9B, and 27B sizes
+  * Expand Flint support to Linux and Windows
+  * Report generation failures in streaming OpenAI Responses
+  * Pass cloud model context lengths to coding tools
+  * Fix Anthropic web search thinking in follow-ups
+- Update generated using pbuild-ai
+
+-------------------------------------------------------------------
+Thu Oct 22 10:26:49 UTC 2026 - Adrian Schröter <[email protected]>
+
+- Updated to version 0.40.8
+  * Add Recursive Flint coding model to Ollama
+  * Apache 2.0-licensed, 40B-parameter MoE with ~2.5B active
+  * Reasoning, multi-turn tool use, and 262K-token context
+  * Reports 54.9% Terminal-Bench 2.1, 74.2% SWE-bench Verified
+  * macOS support enabled now; other GPU support later
+- Update generated using pbuild-ai
+
+-------------------------------------------------------------------
+Wed Oct 21 07:06:38 UTC 2026 - Adrian Schröter <[email protected]>
+
+- Updated to version 0.40.7
+  * Added pplx-decider: a 27B Perplexity decision model fine-tuned
+  from Qwen3.8-27B
+  * Fixed OpenAI-compatible Responses conversations failing when
+  switching between models with and without thinking, or handling
+  compaction items from other providers
+  * Fixed streaming Responses requests where client tool calls (e.g.
+  Codex web search) could emit before open reasoning items, causing
+  misordered/duplicate events
+  * Cancelled remote model requests on ollama.com when the client
+  disconnects, instead of leaving them to run
+  * Enhanced ollama show to report renderer, parser, and chat
+  source; API now exposes these as new fields
+  * Expanded speculative decoding support: llama-server now infers
+  draft types automatically without Ollama identification
+  * Clamped CLAUDE_CODE_AUTO_COMPACT_WINDOW in ollama launch claude
+  to Claude Code's supported range, preventing window size warnings
+  * Updated llama.cpp and MLX
+- Update generated using pbuild-ai
+
+-------------------------------------------------------------------

Old:
----
  ollama-0.40.6.tar.gz

New:
----
  ollama-0.40.9.tar.gz

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Other differences:
------------------
++++++ ollama.spec ++++++
--- /var/tmp/diff_new_pack.8WChHW/_old  2026-10-23 20:30:20.619728666 +0200
+++ /var/tmp/diff_new_pack.8WChHW/_new  2026-10-23 20:30:20.620728708 +0200
@@ -23,7 +23,7 @@
 %global jobs 4
 %endif
 
-%global llama_cpp_version b11669
+%global llama_cpp_version b11803
 %global mlx_version bebd6422f8db88d56e6220520a35bcd146f823f8
 %global mlx_c_version a341b4925024b88b2c593468f16e12f5e17315da
 
@@ -33,6 +33,11 @@
 %else
 %{bcond_with vulkan}
 %endif
+%if 0%{?suse_version} >= 1699
+%{bcond_without rocm}
+%else
+%{bcond_with rocm}
+%endif
 
 %global flavor @BUILD_FLAVOR@%{nil}
 %global preset_flavor %flavor
@@ -40,7 +45,7 @@
 %if "%{flavor}" == "cuda"
 %global preset_flavor llama_cuda_v13_linux
 %if %{with cuda}
-ExclusiveArch:  x86_64
+ExclusiveArch:  aarch64 x86_64
 %else
 ExclusiveArch:  SKIP_IT
 %endif
@@ -50,10 +55,10 @@
 # ROCm is only available for x86_64
 ExclusiveArch:  x86_64
 %global preset_flavor rocm_v7_2_linux
-%if 0%{?suse_version} >= 1699
-%{bcond_without rocm}
+%if %{with rocm}
+ExclusiveArch:  x86_64
 %else
-%{bcond_with rocm}
+ExclusiveArch:  SKIP_IT
 %endif
 %endif
 
@@ -77,7 +82,7 @@
 %define cuda_version %{cuda_version_major}-%{cuda_version_minor}
 
 Name:           ollama
-Version:        0.40.6
+Version:        0.40.9
 Release:        0
 Summary:        Tool for running AI models on-premise
 License:        MIT
@@ -127,8 +132,10 @@
 BuildRequires:  ollama-vulkan
 %endif
 %if %{with rocm}
+%ifarch x86_64
 BuildRequires:  ollama-rocm
 %endif
+%endif
 %if %{with cuda}
 BuildRequires:  ollama-cuda
 %endif

++++++ _agents.obscpio ++++++

++++++ _scmsync.obsinfo ++++++
--- /var/tmp/diff_new_pack.8WChHW/_old  2026-10-23 20:30:20.780735423 +0200
+++ /var/tmp/diff_new_pack.8WChHW/_new  2026-10-23 20:30:20.784735590 +0200
@@ -1,5 +1,5 @@
-mtime: 1792393520
-commit: 1e9128b8d7caceb9777f7e250df3a485ce5ec2ef28989cde08bc17cc7e647c0f
+mtime: 1792742322
+commit: e4bad929b2a305200f186e67cc85e3f2a725aaa4218f28e96f9fcbe5f37cb037
 url: https://src.opensuse.org/AI/ollama.git
 revision: main
 

++++++ build.specials.obscpio ++++++

++++++ build.specials.obscpio ++++++
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn' 
'--exclude=.svnignore' old/.gitignore new/.gitignore
--- old/.gitignore      1970-01-01 01:00:00.000000000 +0100
+++ new/.gitignore      2026-10-23 09:58:42.000000000 +0200
@@ -0,0 +1,17 @@
+*.obscpio
+*.osc
+_build.*
+.pbuild
+.assets
+ollama
+llama.cpp
+mlx
+mlx-c
+
+*.obsinfo
+binaries
+_service:*
+*.orig
+*.rej
+*~
+osc-collab.*

++++++ llama.cpp-main.tar.xz ++++++
/work/SRC/openSUSE:Factory/ollama/llama.cpp-main.tar.xz 
/work/SRC/openSUSE:Factory/.ollama.new.2252680/llama.cpp-main.tar.xz differ: 
char 15, line 1

++++++ mlx-c-main.tar.xz ++++++

++++++ mlx-main.tar.xz ++++++

++++++ ollama-0.40.6.tar.gz -> ollama-0.40.9.tar.gz ++++++
/work/SRC/openSUSE:Factory/ollama/ollama-0.40.6.tar.gz 
/work/SRC/openSUSE:Factory/.ollama.new.2252680/ollama-0.40.9.tar.gz differ: 
char 19, line 1

++++++ vendor.tar.gz ++++++
/work/SRC/openSUSE:Factory/ollama/vendor.tar.gz 
/work/SRC/openSUSE:Factory/.ollama.new.2252680/vendor.tar.gz differ: char 14, 
line 1

Reply via email to