From ec19771b8ba83cb370a410670c0485e232483a3b Mon Sep 17 00:00:00 2001
From: Frank <frank@dankware>
Date: Wed, 21 Oct 2026 02:25:14 -0700
Subject: [PATCH] docs(llama-cpp): deployment detail goes back to the setup
 page
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The overview page carried `docker run` invocations (including a CUDA variant, on a fleet with no
NVIDIA card left in it) and the `LLAMA_ARG_` environment rule, while `setup.md` — the page whose
own intro promises "build instructions, Docker, and systemd deployment" — carried the Compose
example and the same env form in use. Two places to look, one place to be wrong.

`setup.md` now states the env-prefix rule where the flags live (and why units prefer it: one
flag per line is auditable, a forty-flag command line is not), and the overview keeps only the
API surface and the GGUF format, pointing at the sections that absorbed the rest.

Refs #64
---
 omp-docs/tools/llama-cpp/index.md | 29 +++--------------------------
 omp-docs/tools/llama-cpp/setup.md |  8 ++++++++
 2 files changed, 11 insertions(+), 26 deletions(-)

diff --git a/omp-docs/tools/llama-cpp/index.md b/omp-docs/tools/llama-cpp/index.md
index 8425263..b086ec3 100644
--- a/omp-docs/tools/llama-cpp/index.md
+++ b/omp-docs/tools/llama-cpp/index.md
@@ -124,31 +124,8 @@ print(completion.choices[0].message.content)
 
 ---
 
-## Docker
-
-```bash
-docker run -p 8080:8080 -v /models:/models \
-  ghcr.io/ggml-org/llama.cpp:server \
-  -m /models/model.gguf --host 0.0.0.0 --port 8080
-
-# CUDA variant
-docker run -p 8080:8080 --gpus all \
-  ghcr.io/ggml-org/llama.cpp:server-cuda \
-  -m /models/model.gguf --n-gpu-layers 99
-```
-
----
-
-## Environment Variables
-
-All CLI options can be set via `LLAMA_ARG_` prefix:
-
-```bash
-LLAMA_ARG_MODEL=/models/model.gguf LLAMA_ARG_CTX_SIZE=8192 llama-server
-```
-
----
-
 ## Cross-References
 
-- See [Setup Guide](setup.md) for build instructions, GPU offloading, and deployment.
+- See [Setup Guide](setup.md) for build instructions, GPU offloading, the `LLAMA_ARG_*`
+  environment form, the Docker Compose example, and running under systemd. Deployment lives
+  there; this page is the API surface and the GGUF format.
diff --git a/omp-docs/tools/llama-cpp/setup.md b/omp-docs/tools/llama-cpp/setup.md
index 1bc7b35..b220d9b 100644
--- a/omp-docs/tools/llama-cpp/setup.md
+++ b/omp-docs/tools/llama-cpp/setup.md
@@ -111,6 +111,14 @@ cmake --build build --config Release -t llama-server
 | `-lora FNAME` | LoRA adapter (repeatable) | — |
 | `-ngl, --gpu-layers N` | Layers to offload to GPU | auto |
 
+### Environment Variables
+
+Every CLI option can be set from the environment instead: prefix the upper-snake form of the
+long flag with `LLAMA_ARG_` — `LLAMA_ARG_MODEL=/models/model.gguf LLAMA_ARG_CTX_SIZE=8192
+llama-server`. This is how the container and systemd units pass configuration (see the Compose
+example below), because a unit file with one flag per line is auditable and a shell command line
+with forty is not.
+
 ### Context & Performance
 
 | Flag | Description | Default |