Skip to content

Qwen3.8-Flash-Next on any consumer hardware: one-click install for Windows / Linux. Strata inference engine, OpenAI/Anthropic API on localhost, optional image input.

C++

Docker: keep only gguf-py and the ggml marker of the baked llama.cpp; root Dockerfile PATH fix

Niko12214895b1cfa09f
diff --git a/Dockerfile b/Dockerfile
index e083c21f8..5c825ce91 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -20,7 +20,7 @@ ARG RUNTIME_BASE=ubuntu:24.04
# ---- 1. compile the engine and make its component package
FROM ${CUDA_IMAGE} AS engine-build
-ENV DEBIAN_FRONTEND=noninteractive PYTHONUNBUFFERED=1
+ENV DEBIAN_FRONTEND=noninteractive PYTHONUNBUFFERED=1 PATH=/venv/bin:$PATH
RUN apt-get update && apt-get install -y --no-install-recommends \
build-essential ca-certificates git libgomp1 libatomic1 python3 python3-pip python3-venv unzip \
&& rm -rf /var/lib/apt/lists/*
@@ -69,10 +69,15 @@ RUN python3 /src/packaging/package_terminal.py --out /out --cache /cache \
&& tar -xzf /out/strata-terminal-*-linux-x86_64.tar.gz -C /opt/strata \
&& mv /opt/strata/strata-terminal-* /opt/strata/terminal \
&& /opt/strata/terminal/bin/strata --version
-# llama.cpp at the commit setup.py pins (its vocabularies and gguf tools): fetched here, so the first start does not depend on GitHub
+# llama.cpp at the commit setup.py pins: fetched here, so the first start does not depend on GitHub. Only gguf-py (the model preparation
+# tools) and the file setup.py looks for (ggml/CMakeLists.txt) are kept: 170 MB become a few
RUN cd /opt/strata/terminal && STRATA_CONTROL_INSTALL=1 STRATA_INSTALL_DIR=/opt/strata/seed PYTHONNOUSERSITE=1 \
/opt/strata/terminal/python/bin/python3 -c "import setup; print(setup.get_llama_cpp())" \
- && rm -rf /opt/strata/seed/third_party/*.zip* && du -sh /opt/strata/seed
+ && rm -rf /opt/strata/seed/third_party/*.zip* \
+ && cd /opt/strata/seed/third_party/llama.cpp \
+ && find . -mindepth 1 -maxdepth 1 ! -name gguf-py ! -name ggml ! -name 'LICENSE*' -exec rm -rf {} + \
+ && find ggml -mindepth 1 -depth ! -type d ! -name CMakeLists.txt -delete && find ggml -mindepth 1 -type d -empty -delete \
+ && du -sh /opt/strata/seed
# ---- 3. the runtime image (the same as docker/Dockerfile's, FLAVOR=cuda)
FROM ${RUNTIME_BASE}
diff --git a/docker/Dockerfile b/docker/Dockerfile
index 1e3f27dce..78bb53867 100644
--- a/docker/Dockerfile
+++ b/docker/Dockerfile
@@ -36,10 +36,15 @@ RUN python3 /src/packaging/package_terminal.py --out /out --cache /cache \
&& tar -xzf /out/strata-terminal-*-linux-x86_64.tar.gz -C /opt/strata \
&& mv /opt/strata/strata-terminal-* /opt/strata/terminal \
&& /opt/strata/terminal/bin/strata --version
-# llama.cpp at the commit setup.py pins (its vocabularies and gguf tools): fetched here, so the first start does not depend on GitHub
+# llama.cpp at the commit setup.py pins: fetched here, so the first start does not depend on GitHub. Only gguf-py (the model preparation
+# tools) and the file setup.py looks for (ggml/CMakeLists.txt) are kept: 170 MB become a few
RUN cd /opt/strata/terminal && STRATA_CONTROL_INSTALL=1 STRATA_INSTALL_DIR=/opt/strata/seed PYTHONNOUSERSITE=1 \
/opt/strata/terminal/python/bin/python3 -c "import setup; print(setup.get_llama_cpp())" \
- && rm -rf /opt/strata/seed/third_party/*.zip* && du -sh /opt/strata/seed
+ && rm -rf /opt/strata/seed/third_party/*.zip* \
+ && cd /opt/strata/seed/third_party/llama.cpp \
+ && find . -mindepth 1 -maxdepth 1 ! -name gguf-py ! -name ggml ! -name 'LICENSE*' -exec rm -rf {} + \
+ && find ggml -mindepth 1 -depth ! -type d ! -name CMakeLists.txt -delete && find ggml -mindepth 1 -type d -empty -delete \
+ && du -sh /opt/strata/seed
# ---- 2. the engine package, checked before it is allowed into the image: every file's hash, and every library it imports is
# bundled or belongs to the OS / GPU driver (packaging/binary_deps.py, the check the packager runs)