Leonid Pershin
91d2ce0fab
- Added new environment variables in `env.example` and `gpu-rent.vars.example` for fine-tuning LLM settings, including CUDA build options and model version pinning. - Updated `llm.md` documentation to include detailed descriptions of new configuration options and usage cases for LLM setups. - Enhanced the `provision.py` script to forward new environment variables during remote installations, improving the installation process for LLM components. - Modified the `install_llamacpp.sh` script to support conditional CUDA builds and asset URL overrides, ensuring better compatibility with various environments. - Improved logging in the installation scripts to provide clearer feedback during the setup process.
85 lines
2.6 KiB
Plaintext
85 lines
2.6 KiB
Plaintext
# Copy to .env in this repo (never commit the copy)
|
||
# How to fill: docs/setup.md
|
||
# X-Token from the panel does NOT work here.
|
||
|
||
OS_AUTH_URL=https://cloud.api.selcloud.ru/identity/v3
|
||
OS_USER_DOMAIN_NAME=
|
||
OS_USERNAME=
|
||
OS_PASSWORD=
|
||
OS_PROJECT_ID=
|
||
OS_REGION_NAME=ru-7
|
||
GPU_RENT_AZ=ru-7a
|
||
|
||
SSH_PRIVATE_KEY_PATH=
|
||
SSH_USER=ubuntu
|
||
# Prefer auto public /32 (leave unset). Open to world only for spike / WARP:
|
||
# GPU_RENT_SSH_CIDR=0.0.0.0/0
|
||
GPU_RENT_SSH_CIDR=
|
||
|
||
BOOT_VOLUME_ID=
|
||
DATA_VOLUME_ID=
|
||
DATA_VOLUME_SIZE_GB=100
|
||
BOOT_SNAPSHOT_NAME=gpu-rent-boot-ok
|
||
|
||
CIVITAI_API_TOKEN=
|
||
CIVITAI_API_HOST=civitai.red
|
||
MODELS_MANIFEST=
|
||
LOCAL_MODELS_DIR=
|
||
LOCAL_WILDCARDS_DIR=
|
||
LOCAL_WORKFLOWS_DIR=
|
||
LOCAL_OUTPUT_DIR=
|
||
EXTENSIONS_MANIFEST=
|
||
GIT_TOKEN=
|
||
|
||
AUTOCOMPLETE_ENABLED=true
|
||
AUTOCOMPLETE_GITHUB_REPO=DominikDoom/a1111-sd-webui-tagcomplete
|
||
AUTOCOMPLETE_GITHUB_PATH=tags/danbooru.csv
|
||
AUTOCOMPLETE_GITHUB_REF=main
|
||
AUTOCOMPLETE_FILENAME=danbooru.csv
|
||
|
||
SWARMUI_LOCAL_PORT=17801
|
||
# Optional LLM beside SwarmUI: none | ollama | llamacpp (setup / --ollama / --llamacpp)
|
||
# ENABLE_SWARMUI=true
|
||
# WORKLOAD=llm # llm-only (no SwarmUI); requires LLM_RUNTIME≠none
|
||
LLM_RUNTIME=none
|
||
OLLAMA_LOCAL_PORT=17811
|
||
LLAMACPP_LOCAL_PORT=17812
|
||
# OLLAMA_MODELS_MANIFEST=
|
||
# LLAMACPP_MODELS_MANIFEST=
|
||
# Pin / тонкая настройка LLM (несecреты; удобнее в gpu-rent.vars — см. кейсы A–H):
|
||
# LLAMACPP_TAG=b10545
|
||
# LLAMACPP_BUILD_CUDA=1
|
||
# LLAMACPP_FORCE_REINSTALL=1
|
||
# LLAMACPP_ASSET_URL=
|
||
# LLAMACPP_SHA256=
|
||
# LLAMACPP_NGL=40
|
||
# LLAMACPP_CTX=4096
|
||
# LLAMACPP_EXTRA_ARGS=--flash-attn on
|
||
# OLLAMA_VERSION=0.6.5
|
||
# OLLAMA_SHA256=
|
||
# CIVITAI_API_TOKEN=
|
||
# CIVITAI_API_HOST=civitai.red
|
||
# Hugging Face (GGUF / gated HF URLs / capture fallback metadata):
|
||
# HF_TOKEN= # or HUGGING_FACE_HUB_TOKEN — https://huggingface.co/settings/tokens
|
||
# git pull SwarmUI + extensions on each up (default true). CLI: --no-update
|
||
UPDATE_GIT=true
|
||
|
||
DEFAULT_FLAVOR_ID=
|
||
FLAVOR_PREFERENCE=4090-24,4090-48,a5000,a100-40
|
||
FLAVOR_FALLBACK=true
|
||
# Pools to probe before pick (ru-6 = Selectel multi-zone). Comma-separated.
|
||
SCAN_POOLS=ru-6,ru-7
|
||
DEFAULT_SPOT=true
|
||
KEEP_FLOATING_IP=false
|
||
IDLE_MINUTES=60
|
||
IDLE_GRACE_MINUTES=90
|
||
# Optional local safety net (after: gpu-rent watchdog install)
|
||
# LOCAL_WATCHDOG_GRACE_MINUTES=10
|
||
PULL_OUTPUT=false
|
||
NOTIFY_READY=true
|
||
# Статический API-ключ панели (X-Token) — только для баланса, не для OpenStack:
|
||
# SELECTEL_API_TOKEN=
|
||
# Уведомления о списании (local-watchdog tick), шаг с момента up:
|
||
# BALANCE_NOTIFY_STEP_RUB=200
|
||
# BALANCE_NOTIFY_LOW_RUB=0
|