| Index | index by Group | index by Distribution | index by Vendor | index by creation date | index by Name | Mirrors | Help | Search |
| Name: llama-cpp-server | Distribution: OpenMandriva Lx |
| Version: b10549 | Vendor: OpenMandriva |
| Release: 1 | Build date: Fri Aug 21 18:34:42 2026 |
| Group: Servers | Build host: ryzen9-6.openmandriva.org |
| Size: 0 | Source RPM: llama-cpp-b10549-1.src.rpm |
| Packager: bero <bero@lindev.ch> | |
| Url: https://github.com/ggml-org/llama.cpp | |
| Summary: OpenAI API compatible server for llama-cpp | |
OpenAI API compatible server for llama-cpp.
The server also serves an embedded web UI at http://HOST:PORT/ .
Config: /etc/sysconfig/llama-server
To test (wait until weights are loaded; /v1/chat/completions returns
HTTP 503 {"error":{"message":"Loading model"}} until then — a 6–8B
GGUF on GPU is typically ~1 minute):
until curl -sf http://localhost:8080/health; do sleep 1; done
curl http://localhost:8080/v1/chat/completions \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $(grep ^API_KEY= /etc/sysconfig/llama-server 2>/dev/null |cut -d= -f2-)" \
-d '{"model":"any","messages":[{"role":"user","content":"Hello"}]}'
Authorization must match API_KEY in /etc/sysconfig/llama-server
(omit the header if API_KEY is unset).
Apertus 1.0 (Instruct-2509) and Apertus 1.5 (swiss-ai/Apertus-v1.5-*)
are both supported. 1.5 needs a matching mmproj GGUF for image/audio.
MIT AND Apache-2.0 AND LicenseRef-Fedora-Public-Domain
/etc/sysconfig/llama-server /usr/bin/llama-server /usr/lib/systemd/system/llama.service /usr/lib64/libllama-server-impl.so
Generated by rpm2html 1.8.1
Fabrice Bellet, Fri Aug 21 22:00:40 2026