noonghunnaandClaude Opus 4.8
8ffa46d51d
#358 wired the registry/profile/patches but missed setup.sh's own hardcoded per-model dispatch, so `setup.sh diffusiongemma-26b-a4b` errored 'unsupported model'. Add the model_label, the main weights dispatch (PRIMARY_WEIGHT_KEY= diffusiongemma-26b-a4b:fp8 — fp8-only, default WEIGHTS=autoround is a no-op), the Supported: list, and a next-steps hint (switch.sh --force / gpu-mode dgemma). Verified: weights.py entry diffusiongemma-26b-a4b:fp8 → RedHatAI/diffusiongemma-26B-A4B-it-FP8-dynamic; test-model-weights-registry green. Completes the 'users can use setup' wiring. Co-Authored-By: Claude Opus 4.8 <[email protected]>
697 lines
28 KiB
Bash
Executable File
697 lines
28 KiB
Bash
Executable File
#!/usr/bin/env bash
|
|
#
|
|
# Model-aware one-shot setup for club-3090.
|
|
#
|
|
# bash scripts/setup.sh # interactive model picker in a TTY
|
|
# bash scripts/setup.sh <model-name> # scripted/CI positional form
|
|
#
|
|
# Currently supported model families are listed in usage(). Exact repositories,
|
|
# files, and local subdirectories live in scripts/lib/profiles/models/*.yml.
|
|
#
|
|
# What it does (per supported model):
|
|
# - clones Sandermage/genesis-vllm-patches into models/<model>/vllm/patches/genesis
|
|
# (vLLM-only; skip with SKIP_GENESIS=1 if you only need llama.cpp / SGLang;
|
|
# Gemma 4 doesn't fetch Genesis at all yet)
|
|
# - downloads model weights into $MODEL_DIR with SHA256 verification
|
|
# against HF x-linked-etag
|
|
# - downloads the always-required drafter (Gemma 4: MTP "assistant"; Qwen3.6:
|
|
# no always-required drafter — DFlash is optional via WITH_DFLASH_DRAFT=1)
|
|
#
|
|
# Env vars (optional):
|
|
# MODEL_DIR Where to place model weights. Default: <repo>/models-cache
|
|
# WEIGHTS 'autoround' (default, vLLM INT4) or 'gguf' (llama.cpp /
|
|
# ik_llama). gguf fetches the Q4_K_M MTP GGUF + mmproj for
|
|
# the llamacpp/* + ik-llama/* composes (qwen3.6-27b only),
|
|
# and skips Genesis. Use this if you're serving via
|
|
# llama.cpp/ik_llama rather than vLLM.
|
|
# HF_TOKEN HF token (public models, usually unnecessary)
|
|
# SKIP_MODEL Set to 1 to skip the model download step
|
|
# SKIP_GENESIS Set to 1 to skip cloning Genesis patches
|
|
# WITH_DFLASH_DRAFT Set to 1 to ALSO download the model family's DFlash
|
|
# drafter when one is registered in profiles/models/*.yml.
|
|
# Default: 0.
|
|
# Note: draft model is still under training as of
|
|
# 2026-04-26; bench numbers in DUAL_CARD.md were
|
|
# measured against that snapshot. AL improvements
|
|
# expected when z-lab tags training-complete.
|
|
# PREFLIGHT_DISK_GB Required free space at MODEL_DIR (default: 25, or
|
|
# 28 if WITH_DFLASH_DRAFT=1)
|
|
#
|
|
# Idempotent: safe to re-run — skips steps already done.
|
|
|
|
set -euo pipefail
|
|
|
|
ROOT_DIR="$(cd -- "$(dirname -- "${BASH_SOURCE[0]}")/.." && pwd)"
|
|
WEIGHTS_READER="${ROOT_DIR}/scripts/lib/profiles/weights.py"
|
|
|
|
usage() {
|
|
echo "Usage: $0 <model-name>"
|
|
echo " $0 # interactive model picker in a TTY"
|
|
echo ""
|
|
echo "Run with no model name in a normal terminal to open the hardware-aware"
|
|
echo "model picker. Use the positional form in scripts/CI to skip prompts."
|
|
echo ""
|
|
echo "Supported model names:"
|
|
echo " qwen3.6-27b"
|
|
echo " qwen3.6-35b-a3b"
|
|
echo " gemma-4-31b"
|
|
echo " gemma-4-26b-a4b"
|
|
echo ""
|
|
echo "Exact catalog entry fetch: WEIGHT_KEY=<registry-key> $0 <model-name>"
|
|
}
|
|
|
|
model_label() {
|
|
case "$1" in
|
|
qwen3.6-27b) echo "Qwen 3.6 27B" ;;
|
|
qwen3.6-35b-a3b) echo "Qwen 3.6 35B-A3B" ;;
|
|
gemma-4-31b) echo "Gemma 4 31B" ;;
|
|
gemma-4-26b-a4b) echo "Gemma 4 26B-A4B" ;;
|
|
diffusiongemma-26b-a4b) echo "DiffusionGemma 26B-A4B (dLLM)" ;;
|
|
*) echo "$1" ;;
|
|
esac
|
|
}
|
|
|
|
load_weight_recipe() {
|
|
local key="$1"
|
|
local env_lines
|
|
command -v python3 >/dev/null 2>&1 || {
|
|
echo "ERROR: python3 is required to read profile weight recipes." >&2
|
|
exit 1
|
|
}
|
|
env_lines="$(python3 "${WEIGHTS_READER}" entry "$key" 2>/dev/null)" || {
|
|
echo "ERROR: could not resolve weight recipe '${key}' from scripts/lib/profiles/models/*.yml." >&2
|
|
echo " Install python3-yaml/PyYAML if missing, or check the catalog key." >&2
|
|
exit 1
|
|
}
|
|
eval "$env_lines"
|
|
if [[ -n "${WEIGHT_MODEL:-}" && "${WEIGHT_MODEL}" != "${MODEL_NAME}" ]]; then
|
|
echo "ERROR: weight recipe '${key}' belongs to ${WEIGHT_MODEL}, not ${MODEL_NAME}." >&2
|
|
exit 1
|
|
fi
|
|
if [[ -z "${WEIGHT_REPO:-}" ]]; then
|
|
echo "ERROR: weight recipe '${key}' has no direct download recipe." >&2
|
|
[[ -n "${WEIGHT_MANUAL_NOTE:-}" ]] && echo " ${WEIGHT_MANUAL_NOTE}" >&2
|
|
exit 1
|
|
fi
|
|
MODEL_REPO="${WEIGHT_REPO}"
|
|
MODEL_SUBDIR="${WEIGHT_SUBDIR}"
|
|
GGUF_FILES="${WEIGHT_FILES}"
|
|
VERIFY_GLOB="${WEIGHT_VERIFY_GLOB:-*.safetensors}"
|
|
echo "[model] ${WEIGHT_KEY} -> ${MODEL_REPO} ${WEIGHT_FILES} -> ${MODEL_SUBDIR}"
|
|
}
|
|
|
|
model_picker_line() {
|
|
local idx="$1" model="$2" size="$3" status mark reason
|
|
status="$(compose_hw_model_status "$ROOT_DIR" "$model" 2>/dev/null || true)"
|
|
reason="${status#*|}"
|
|
if [[ "$status" == ok\|* ]]; then
|
|
mark="✓"
|
|
else
|
|
mark="✗"
|
|
fi
|
|
printf " %s. %-14s (%s) %s %s\n" "$idx" "$(model_label "$model")" "$size" "$mark" "$reason"
|
|
}
|
|
|
|
pick_model_interactive() {
|
|
# shellcheck source=lib/compose-meta.sh
|
|
source "${ROOT_DIR}/scripts/lib/compose-meta.sh"
|
|
|
|
echo "[setup] Which model to download?" >&2
|
|
echo "" >&2
|
|
model_picker_line "1" "qwen3.6-27b" "~14 GB AutoRound INT4" >&2
|
|
model_picker_line "2" "gemma-4-31b" "~21 GB AutoRound INT4 + drafter" >&2
|
|
echo " 3. Both (~30 GB total) downloads both model families" >&2
|
|
echo "" >&2
|
|
while true; do
|
|
local pick
|
|
read -rp "Choice [1-3]: " pick
|
|
case "$pick" in
|
|
1) echo "qwen3.6-27b"; return ;;
|
|
2) echo "gemma-4-31b"; return ;;
|
|
3) echo "both"; return ;;
|
|
*) echo " ! invalid — pick 1, 2, or 3" >&2 ;;
|
|
esac
|
|
done
|
|
}
|
|
|
|
# ---------- Model dispatch ----------
|
|
case "${1:-}" in
|
|
-h|--help)
|
|
usage
|
|
exit 0
|
|
;;
|
|
esac
|
|
|
|
# setup.sh takes a SINGLE positional (the model). It DOWNLOADS WEIGHTS — it does
|
|
# NOT select or boot a serving config. A stray second arg (commonly a launch slug
|
|
# like `vllm/int8`) used to be silently ignored, which let users believe it did
|
|
# something — see issue #250, where the slug was dropped and the real failure
|
|
# (a purged-nightly compose pin) got mis-attributed to it. Reject it loudly and
|
|
# point at the launch path. (`both` recurses with a single arg, so it's unaffected.)
|
|
if [[ $# -gt 1 ]]; then
|
|
echo "ERROR: setup.sh takes a single model name; got extra argument(s): ${*:2}" >&2
|
|
echo " setup.sh only DOWNLOADS WEIGHTS for a model — e.g. bash scripts/setup.sh ${1}" >&2
|
|
echo " To LAUNCH a serving config (a slug such as 'vllm/gemma-int8-mtp'), use:" >&2
|
|
echo " bash scripts/launch.sh --variant <slug> # or: bash scripts/switch.sh <slug>" >&2
|
|
echo " See the slugs available for a model: bash scripts/switch.sh --list" >&2
|
|
exit 64
|
|
fi
|
|
|
|
MODEL_NAME="${1:-}"
|
|
if [[ -z "${MODEL_NAME}" ]]; then
|
|
if [[ -t 0 && -t 1 ]]; then
|
|
MODEL_NAME="$(pick_model_interactive)"
|
|
else
|
|
usage
|
|
echo ""
|
|
echo "(Interactive picker available in a TTY shell. Use the positional form in scripts/CI.)"
|
|
exit 1
|
|
fi
|
|
fi
|
|
|
|
if [[ "${MODEL_NAME}" == "both" ]]; then
|
|
# Resolve MODEL_DIR once in the parent by reusing the normal prompt below,
|
|
# then recurse through the positional form for each model.
|
|
SETUP_BOTH_MODE=1
|
|
MODEL_NAME="qwen3.6-27b"
|
|
else
|
|
SETUP_BOTH_MODE=0
|
|
fi
|
|
|
|
# The profile YAMLs are the only source of download recipes. setup.sh only
|
|
# maps friendly setup knobs (MODEL_NAME, WEIGHTS, WITH_*) to profile keys.
|
|
ALWAYS_DRAFT_KEY=""
|
|
DFLASH_KEY=""
|
|
PRIMARY_WEIGHT_KEY=""
|
|
EXTRA_WEIGHT_KEYS=()
|
|
NEEDS_GENESIS=0
|
|
|
|
case "${MODEL_NAME}" in
|
|
qwen3.6-27b)
|
|
PRIMARY_WEIGHT_KEY="qwen3.6-27b:autoround-int4"
|
|
DFLASH_KEY="qwen3.6-27b:dflash"
|
|
NEEDS_GENESIS=1
|
|
;;
|
|
qwen3.6-35b-a3b)
|
|
PRIMARY_WEIGHT_KEY="qwen3.6-35b-a3b:autoround-int4"
|
|
;;
|
|
gemma-4-31b)
|
|
PRIMARY_WEIGHT_KEY="gemma-4-31b:autoround-int4"
|
|
ALWAYS_DRAFT_KEY="gemma-4-31b:assistant"
|
|
DFLASH_KEY="gemma-4-31b:dflash"
|
|
;;
|
|
gemma-4-26b-a4b)
|
|
PRIMARY_WEIGHT_KEY="gemma-4-26b-a4b:autoround-int4-mixed"
|
|
;;
|
|
diffusiongemma-26b-a4b)
|
|
# dLLM, fp8 only (no autoround variant). Default WEIGHTS=autoround is a no-op
|
|
# here — the fp8 key set below is what's fetched.
|
|
PRIMARY_WEIGHT_KEY="diffusiongemma-26b-a4b:fp8"
|
|
;;
|
|
*)
|
|
echo "ERROR: unsupported model '${MODEL_NAME}'."
|
|
echo "Supported: qwen3.6-27b, qwen3.6-35b-a3b, gemma-4-31b, gemma-4-26b-a4b, diffusiongemma-26b-a4b"
|
|
echo "(To add a new model, extend the model dispatch in scripts/setup.sh and profiles/models/*.yml)"
|
|
exit 1
|
|
;;
|
|
esac
|
|
|
|
# ---------- Weights format / exact registry entry ----------
|
|
# WEIGHTS selects a common weight variant for the model family. WEIGHT_KEY is
|
|
# the exact catalog-entry path used by preflight's fetch-now flow.
|
|
WEIGHTS="${WEIGHTS:-autoround}"
|
|
GGUF_FILES=""
|
|
VERIFY_GLOB="*.safetensors"
|
|
|
|
if [[ -n "${WEIGHT_KEY:-}" ]]; then
|
|
PRIMARY_WEIGHT_KEY="${WEIGHT_KEY}"
|
|
elif [[ "${WEIGHTS}" == "gguf" ]]; then
|
|
case "${MODEL_NAME}" in
|
|
qwen3.6-27b)
|
|
PRIMARY_WEIGHT_KEY="qwen3.6-27b:unsloth-q4km"
|
|
EXTRA_WEIGHT_KEYS+=("qwen3.6-27b:gguf_mmproj_f16")
|
|
NEEDS_GENESIS=0
|
|
;;
|
|
*)
|
|
echo "ERROR: WEIGHTS=gguf is only wired for qwen3.6-27b right now." >&2
|
|
echo " Use WEIGHT_KEY=<model>:<variant> for exact catalog entries." >&2
|
|
exit 1 ;;
|
|
esac
|
|
elif [[ "${WEIGHTS}" == "iq4ks" ]]; then
|
|
case "${MODEL_NAME}" in
|
|
qwen3.6-27b)
|
|
PRIMARY_WEIGHT_KEY="qwen3.6-27b:ubergarm-iq4ks"
|
|
NEEDS_GENESIS=0
|
|
;;
|
|
*)
|
|
echo "ERROR: WEIGHTS=iq4ks is only wired for qwen3.6-27b." >&2
|
|
exit 1 ;;
|
|
esac
|
|
elif [[ "${WEIGHTS}" == "awq" ]]; then
|
|
case "${MODEL_NAME}" in
|
|
gemma-4-31b) PRIMARY_WEIGHT_KEY="gemma-4-31b:awq" ;;
|
|
gemma-4-26b-a4b) PRIMARY_WEIGHT_KEY="gemma-4-26b-a4b:awq" ;;
|
|
*)
|
|
echo "ERROR: WEIGHTS=awq is only wired for gemma-4-31b and gemma-4-26b-a4b." >&2
|
|
exit 1 ;;
|
|
esac
|
|
elif [[ "${WEIGHTS}" != "autoround" ]]; then
|
|
echo "ERROR: WEIGHTS='${WEIGHTS}' not recognized (use 'autoround', 'awq', 'gguf', or 'iq4ks')." >&2
|
|
exit 1
|
|
fi
|
|
|
|
if [[ "${WITH_ASSISTANT_DRAFT:-0}" == "1" ]]; then
|
|
case "${MODEL_NAME}" in
|
|
gemma-4-31b) ALWAYS_DRAFT_KEY="gemma-4-31b:assistant" ;;
|
|
gemma-4-26b-a4b) ALWAYS_DRAFT_KEY="gemma-4-26b-a4b:assistant" ;;
|
|
*)
|
|
echo "ERROR: WITH_ASSISTANT_DRAFT=1 is only wired for Gemma models." >&2
|
|
exit 1 ;;
|
|
esac
|
|
fi
|
|
|
|
load_weight_recipe "${PRIMARY_WEIGHT_KEY}"
|
|
|
|
# ---------- MODEL_DIR resolution ----------
|
|
# Order of precedence:
|
|
# 1. MODEL_DIR already exported in the calling shell → use as-is
|
|
# 2. .env at repo root sets MODEL_DIR → source it
|
|
# 3. Interactive prompt (only if stdin is a TTY) → ask user
|
|
# 4. Silent fallback to <repo>/models-cache → in-repo default
|
|
#
|
|
# The prompt only fires for fresh users on a TTY who haven't set anything.
|
|
# CI / scripted runs (no TTY) get the silent fallback, preserving prior behavior.
|
|
|
|
# Step 2: source repo-root .env if present (lets a saved choice persist)
|
|
if [[ -z "${MODEL_DIR:-}" && -f "${ROOT_DIR}/.env" ]]; then
|
|
# shellcheck source=/dev/null
|
|
set -a; source "${ROOT_DIR}/.env"; set +a
|
|
fi
|
|
|
|
# Step 3: prompt if still unset + interactive
|
|
if [[ -z "${MODEL_DIR:-}" && -t 0 && -t 1 ]]; then
|
|
echo ""
|
|
echo "Where should I put model weights?"
|
|
echo " Models are large (Qwen3.6-27B AutoRound: ~14 GB; Gemma 4 31B: ~21 GB)."
|
|
echo " This dir lives outside the git tree — pick a location with sufficient free space."
|
|
echo ""
|
|
echo " 1) ${ROOT_DIR}/models-cache (in-repo, default — pollutes git tree)"
|
|
echo " 2) ${HOME}/models (recommended for cross-rig — outside repo)"
|
|
echo " 3) custom path"
|
|
echo ""
|
|
while true; do
|
|
read -rp "Choice [1-3] (or set MODEL_DIR env var to skip): " pick
|
|
case "${pick}" in
|
|
1) MODEL_DIR="${ROOT_DIR}/models-cache"; break ;;
|
|
2) MODEL_DIR="${HOME}/models"; break ;;
|
|
3)
|
|
read -rp " Enter absolute path: " custom
|
|
if [[ "${custom}" =~ ^/ ]]; then
|
|
MODEL_DIR="${custom}"; break
|
|
else
|
|
echo " ! must be an absolute path (start with /)" >&2
|
|
fi
|
|
;;
|
|
*) echo " ! invalid — pick 1, 2, or 3" >&2 ;;
|
|
esac
|
|
done
|
|
echo ""
|
|
|
|
# Offer to persist the choice so future runs skip the prompt
|
|
read -rp "Save MODEL_DIR=${MODEL_DIR} to .env so we skip this next time? [Y/n]: " save
|
|
if [[ "${save:-y}" =~ ^[Yy]$ || -z "${save:-}" ]]; then
|
|
if [[ -f "${ROOT_DIR}/.env" ]]; then
|
|
# Update existing .env (replace MODEL_DIR= line if present, else append)
|
|
if grep -qE "^MODEL_DIR=" "${ROOT_DIR}/.env"; then
|
|
sed -i "s|^MODEL_DIR=.*|MODEL_DIR=${MODEL_DIR}|" "${ROOT_DIR}/.env"
|
|
else
|
|
echo "MODEL_DIR=${MODEL_DIR}" >> "${ROOT_DIR}/.env"
|
|
fi
|
|
else
|
|
echo "MODEL_DIR=${MODEL_DIR}" > "${ROOT_DIR}/.env"
|
|
fi
|
|
echo " → saved. (.env is gitignored.)"
|
|
else
|
|
echo " → not saved. Set MODEL_DIR=... when re-running, or you'll get this prompt again."
|
|
fi
|
|
echo ""
|
|
fi
|
|
|
|
# Step 4: silent fallback (preserves prior behavior for non-TTY contexts)
|
|
MODEL_DIR="${MODEL_DIR:-${ROOT_DIR}/models-cache}"
|
|
if [[ "${SETUP_BOTH_MODE:-0}" == "1" ]]; then
|
|
export MODEL_DIR
|
|
echo "[setup] downloading both supported models into ${MODEL_DIR}"
|
|
echo ""
|
|
bash "$0" qwen3.6-27b
|
|
echo ""
|
|
bash "$0" gemma-4-31b
|
|
echo ""
|
|
echo "[setup] ✓ Both models downloaded."
|
|
echo "[setup] Next: bash scripts/launch.sh"
|
|
exit 0
|
|
fi
|
|
GENESIS_DIR="${ROOT_DIR}/models/${MODEL_NAME}/vllm/patches/genesis"
|
|
|
|
cd "${ROOT_DIR}"
|
|
|
|
# ---------- Pre-flight checks ----------
|
|
# Catches the common "first-run failures": missing docker, no GPU visible,
|
|
# disk too small for the ~14 GB AutoRound int4 download. Fails fast with
|
|
# actionable hints rather than mid-download or first-boot crash.
|
|
# shellcheck source=preflight.sh
|
|
source "${ROOT_DIR}/scripts/preflight.sh"
|
|
|
|
# Required disk: model is ~14 GB on disk; 25 GB gives buffer for download
|
|
# temp files + safetensors + tokenizer/config. Add ~3 GB if also pulling
|
|
# the DFlash draft (~1.75 GB packed + buffer for download tempfiles).
|
|
if [[ "${WITH_DFLASH_DRAFT:-0}" == "1" ]]; then
|
|
PREFLIGHT_DISK_GB="${PREFLIGHT_DISK_GB:-28}"
|
|
else
|
|
PREFLIGHT_DISK_GB="${PREFLIGHT_DISK_GB:-25}"
|
|
fi
|
|
|
|
echo "[preflight] checking environment..."
|
|
# docker is soft-warn for setup.sh — this script only fetches genesis + models,
|
|
# no docker invocations until you actually `docker compose up` later. Hard-failing
|
|
# blocks non-docker container-runtime users (microk8s / podman / k8s / manual)
|
|
# from running setup at all (club-3090 disc #48). launch.sh keeps the hard check
|
|
# because it actually invokes docker.
|
|
preflight_docker || echo "[preflight] WARN: docker unavailable — setup will continue (genesis + model fetch don't need docker), but you'll need a working container runtime before 'docker compose up'."
|
|
preflight_gpu 1 || exit 1
|
|
preflight_disk "${MODEL_DIR}" "${PREFLIGHT_DISK_GB}" || exit 1
|
|
preflight_hf_token # soft-warn only; downloads will surface the hard failure
|
|
echo "[preflight] ok."
|
|
echo ""
|
|
|
|
# ---------- WSL2 detection — auto-configure .env for known WSL2 boot crash ----------
|
|
# WSL2 + driver 596.36 + vLLM nightly hit a `gptq_marlin_repack` boot crash
|
|
# with `cudaErrorNotReady`. Workaround is `PYTORCH_CUDA_ALLOC_CONF=expandable_segments:False`
|
|
# (PR #84). The compose default is `expandable_segments:True,max_split_size_mb:512`
|
|
# which works on bare-metal Linux but fails on WSL2 — so we auto-create a .env
|
|
# override here on detected WSL2 systems. Cross-rig validated by @timxx (issue #60),
|
|
# @easel, and others. Safe no-op on bare-metal (only runs when /proc/version
|
|
# contains "microsoft").
|
|
COMPOSE_DIR="${ROOT_DIR}/models/${MODEL_NAME}/vllm/compose"
|
|
if [[ -f /proc/version ]] && grep -qi microsoft /proc/version 2>/dev/null; then
|
|
ENV_FILE="${COMPOSE_DIR}/.env"
|
|
if [[ -d "${COMPOSE_DIR}" ]]; then
|
|
if [[ ! -f "${ENV_FILE}" ]]; then
|
|
cat > "${ENV_FILE}" <<'EOF'
|
|
# WSL2 boot-crash workaround — see PR #84 + issue #60.
|
|
# vLLM + WSL2 + driver 596.36 hit `gptq_marlin_repack` cudaErrorNotReady on boot
|
|
# with the default `expandable_segments:True`. This override fixes it.
|
|
# Auto-created by scripts/setup.sh on detected WSL2 systems. Safe to delete
|
|
# on bare-metal Linux (the compose default works there).
|
|
PYTORCH_CUDA_ALLOC_CONF=expandable_segments:False
|
|
EOF
|
|
echo "[wsl2] detected WSL2 — created ${ENV_FILE} with PYTORCH_CUDA_ALLOC_CONF=expandable_segments:False"
|
|
echo "[wsl2] this fixes the known gptq_marlin_repack boot crash on WSL2 + driver ≥596.36 (issue #60)."
|
|
elif ! grep -q "expandable_segments:False" "${ENV_FILE}"; then
|
|
echo "[wsl2] WARN: detected WSL2 but ${ENV_FILE} exists without the expandable_segments:False override."
|
|
echo "[wsl2] If vLLM fails to boot with cudaErrorNotReady, add:"
|
|
echo "[wsl2] PYTORCH_CUDA_ALLOC_CONF=expandable_segments:False"
|
|
echo "[wsl2] See PR #84 / issue #60 for context."
|
|
else
|
|
echo "[wsl2] detected WSL2 — ${ENV_FILE} already has the expandable_segments:False override. ✓"
|
|
fi
|
|
fi
|
|
fi
|
|
|
|
# ---------- Tool checks ----------
|
|
need() {
|
|
command -v "$1" >/dev/null 2>&1 || {
|
|
echo "ERROR: required tool '$1' not found in PATH." >&2
|
|
exit 1
|
|
}
|
|
}
|
|
need git
|
|
need curl
|
|
need sha256sum
|
|
|
|
echo "Setup root: ${ROOT_DIR}"
|
|
echo "Model dir: ${MODEL_DIR}"
|
|
|
|
# ---------- Genesis patches ----------
|
|
# We track Sandermage's tree at HEAD and rely on tagged commits / SHA pinning
|
|
# in the compose files for reproducibility. The repo layout changed substantially
|
|
# between v7.13 (monolithic patch_genesis_unified.py shim) and v7.14 (modular
|
|
# vllm/_genesis package + per-patch env opts). Newer composes mount the package;
|
|
# the legacy compose still references the v7.13 shim.
|
|
# Pin Genesis to the exact commit our published numbers were measured against.
|
|
# Currently pointing at v7.69 dev tip (commit 2db18df, 2026-05-02 PM). Bumped
|
|
# from v7.66 (fc89395) for the v7.69 patch set, which addresses the 3
|
|
# regressions the v7.68 cross-rig retest found ([club-3090#19] and our
|
|
# v7.68-cliff2-test branch summary):
|
|
# - F1 (PN30 part3 drift-marker bug) — fixed via specific marker
|
|
# `[Genesis PN30 v7.68 dst-shaped]` so part3 idempotency check no longer
|
|
# collides with part1+2 markers in the same file.
|
|
# - F2 (P103 setattr lost on `exec vllm serve`) — fixed via self-install
|
|
# hook text-patched into chunk.py end-of-file. Survives any startup
|
|
# mechanism (workers, fork, spawn, exec). The "rebound at 0 caller sites"
|
|
# log message in v7.68 was misleading — internal callers DID get the
|
|
# setattr in the entrypoint shell process, but `exec` replaced the image
|
|
# and lost it. v7.69 hook fires every time chunk.py imports.
|
|
# - F3 (PN32 v1 chunked at wrong level) — rewritten as PN32 v2 to patch
|
|
# `_forward_core` directly + thread initial_state via prior chunk's
|
|
# last_recurrent_state. Composes with P103: v2 chunks the OUTER FLA
|
|
# call, P103 chunks INSIDE the FLA inner h tensor.
|
|
# Recommended Cliff 2 closure env bundle for single-24GB-GPU:
|
|
# GENESIS_ENABLE_P103=1 (close inner FLA h tensor)
|
|
# GENESIS_ENABLE_PN32_GDN_CHUNKED_PREFILL=1 (close outer FLA call buffer)
|
|
# GENESIS_PN32_GDN_CHUNK_SIZE=8192 (default)
|
|
# GENESIS_PN32_GDN_CHUNK_THRESHOLD=16384 (default)
|
|
# GENESIS_FLA_FWD_H_MAX_T=16384 (P103 default)
|
|
# v7.69 also retains v7.68's accept-and-fold of our 3 cross-rig sidecars:
|
|
# PN25 v7.68 (TP=1 worker-spawn registration), PN30 v7.68 (DS conv-state
|
|
# layout dst-shaped temp), PN34 (vllm#39226 runtime workspace_lock).
|
|
# Pinned to dev SHA 2db18df because v7.69 is feature-complete on dev pending
|
|
# our retest validation; if clean, Sander will tag stable.
|
|
# Bumping GENESIS_PIN requires re-running verify-stress.sh against your composes
|
|
# to confirm the new commit works on your config.
|
|
GENESIS_PIN="${GENESIS_PIN:-7b9fd319}"
|
|
|
|
if [[ "${NEEDS_GENESIS:-1}" != "1" ]]; then
|
|
echo "[genesis] ${MODEL_NAME} doesn't use Genesis — skipping clone."
|
|
elif [[ "${SKIP_GENESIS:-0}" != "1" ]]; then
|
|
if [[ -d "${GENESIS_DIR}/.git" ]]; then
|
|
echo "[genesis] Already cloned at ${GENESIS_DIR} — fetching + checking out ${GENESIS_PIN} ..."
|
|
(cd "${GENESIS_DIR}" && git fetch origin && git checkout "${GENESIS_PIN}" 2>&1 | tail -3)
|
|
else
|
|
echo "[genesis] Cloning Sandermage/genesis-vllm-patches at ${GENESIS_PIN} ..."
|
|
# Full clone (commit SHAs aren't reachable via --branch + --depth 1).
|
|
git clone https://github.com/Sandermage/genesis-vllm-patches.git "${GENESIS_DIR}"
|
|
(cd "${GENESIS_DIR}" && git checkout "${GENESIS_PIN}")
|
|
fi
|
|
|
|
# v7.14+ layout sanity check
|
|
if [[ ! -d "${GENESIS_DIR}/vllm/_genesis" ]]; then
|
|
echo "ERROR: genesis tree at ${GENESIS_PIN} missing vllm/_genesis package." >&2
|
|
echo " Re-run with GENESIS_PIN=<other-ref> to try a different version." >&2
|
|
exit 1
|
|
fi
|
|
echo "[genesis] Pinned to ${GENESIS_PIN} ($(cd "${GENESIS_DIR}" && git rev-parse --short HEAD))"
|
|
|
|
# v7.69 ships PN25 + PN30 + PN34 directly (Sander's accept-and-fold of our
|
|
# cross-rig sidecars). Local patch_pn25_genesis_register_fix.py +
|
|
# patch_pn30_dst_shaped_temp_fix.py + patch_workspace_lock_disable.py are
|
|
# now redundant. Sidecar Python files retained in vllm/patches/ for
|
|
# rollback if any v7.69 patch regresses on your config.
|
|
else
|
|
echo "[genesis] SKIP_GENESIS=1 — not cloning."
|
|
fi
|
|
|
|
# ---------- Model download ----------
|
|
if [[ "${SKIP_MODEL:-0}" == "1" ]]; then
|
|
echo "[model] SKIP_MODEL=1 — not downloading."
|
|
exit 0
|
|
fi
|
|
|
|
_hf_download_repo() {
|
|
local repo="$1"
|
|
local subdir="$2"
|
|
local files="${3:-}"
|
|
mkdir -p "${MODEL_DIR}/${subdir}"
|
|
if command -v hf >/dev/null 2>&1; then
|
|
echo "[model] Using 'hf download' (hf_transfer if available) ..."
|
|
# files is intentionally word-split: empty -> whole repo; non-empty -> selected files.
|
|
HF_HUB_ENABLE_HF_TRANSFER=1 HF_HUB_DISABLE_XET=1 \
|
|
hf download "$repo" ${files} --local-dir "${MODEL_DIR}/${subdir}"
|
|
elif command -v huggingface-cli >/dev/null 2>&1; then
|
|
echo "[model] Using 'huggingface-cli download' ..."
|
|
HF_HUB_ENABLE_HF_TRANSFER=1 HF_HUB_DISABLE_XET=1 \
|
|
huggingface-cli download "$repo" ${files} --local-dir "${MODEL_DIR}/${subdir}"
|
|
else
|
|
echo "ERROR: neither 'hf' nor 'huggingface-cli' found. Install with:" >&2
|
|
echo " pip install 'huggingface-hub[hf_transfer]'" >&2
|
|
echo "or:" >&2
|
|
echo " uv tool install --with hf_transfer huggingface-hub" >&2
|
|
exit 1
|
|
fi
|
|
}
|
|
|
|
_verify_downloaded_files() {
|
|
local repo="$1"
|
|
local subdir="$2"
|
|
local verify_glob="$3"
|
|
local fail=0 count=0 f expected actual
|
|
|
|
echo "[verify] Checking SHA256 of every ${verify_glob} against HF x-linked-etag ..."
|
|
cd "${MODEL_DIR}/${subdir}"
|
|
for f in ${verify_glob}; do
|
|
[[ -f "$f" ]] || continue
|
|
count=$((count + 1))
|
|
expected="$(curl -sfI "https://huggingface.co/${repo}/resolve/main/$f" \
|
|
| grep -i '^x-linked-etag:' | tr -d '"\r' | awk '{print $NF}' || true)"
|
|
actual="$(sha256sum "$f" | awk '{print $1}')"
|
|
if [[ -z "$expected" ]]; then
|
|
printf " %-50s SKIP (no etag)\n" "$f"
|
|
elif [[ "$expected" == "$actual" ]]; then
|
|
printf " %-50s OK\n" "$f"
|
|
else
|
|
printf " %-50s FAIL exp=%.12s act=%.12s\n" "$f" "$expected" "$actual"
|
|
fail=$((fail + 1))
|
|
fi
|
|
done
|
|
cd "${ROOT_DIR}"
|
|
|
|
if [[ "$fail" != "0" ]]; then
|
|
echo "[verify] ${fail} file(s) failed SHA check." >&2
|
|
echo " Delete ${MODEL_DIR}/${subdir} and re-run setup.sh." >&2
|
|
exit 1
|
|
fi
|
|
if [[ "$count" == "0" ]]; then
|
|
echo "[verify] No ${verify_glob} found in ${MODEL_DIR}/${subdir} — download may have failed." >&2
|
|
exit 1
|
|
fi
|
|
echo "[done] ${count} file(s) SHA-verified in ${subdir}."
|
|
}
|
|
|
|
download_weight_key() {
|
|
local key="$1"
|
|
load_weight_recipe "$key"
|
|
echo "[model] Downloading ${WEIGHT_LABEL:-$key} ..."
|
|
_hf_download_repo "$WEIGHT_REPO" "$WEIGHT_SUBDIR" "$WEIGHT_FILES"
|
|
_verify_downloaded_files "$WEIGHT_REPO" "$WEIGHT_SUBDIR" "$WEIGHT_VERIFY_GLOB"
|
|
}
|
|
|
|
VERIFY_GLOB="${VERIFY_GLOB_OVERRIDE:-*.safetensors}"
|
|
_hf_download_repo "${MODEL_REPO}" "${MODEL_SUBDIR}" "${GGUF_FILES}"
|
|
_verify_downloaded_files "${MODEL_REPO}" "${MODEL_SUBDIR}" "${VERIFY_GLOB}"
|
|
|
|
for extra_key in "${EXTRA_WEIGHT_KEYS[@]}"; do
|
|
download_weight_key "$extra_key"
|
|
done
|
|
|
|
echo ""
|
|
[[ -d "${GENESIS_DIR}/.git" ]] && echo " Genesis pinned at ${GENESIS_PIN} ($(cd "${GENESIS_DIR}" && git rev-parse --short HEAD))."
|
|
echo ""
|
|
|
|
# ---------- Optional / companion draft models ----------
|
|
if [[ -n "${ALWAYS_DRAFT_KEY:-}" ]] && [[ "${SKIP_MODEL:-0}" != "1" ]]; then
|
|
echo "[draft] downloading required companion drafter ${ALWAYS_DRAFT_KEY} ..."
|
|
download_weight_key "${ALWAYS_DRAFT_KEY}"
|
|
echo ""
|
|
fi
|
|
|
|
if [[ "${WITH_DFLASH_DRAFT:-0}" == "1" ]] && [[ "${SKIP_MODEL:-0}" != "1" ]]; then
|
|
if [[ -z "${DFLASH_KEY:-}" ]]; then
|
|
echo "ERROR: WITH_DFLASH_DRAFT=1 is not wired for ${MODEL_NAME}." >&2
|
|
exit 1
|
|
fi
|
|
echo "[dflash] WITH_DFLASH_DRAFT=1 — downloading ${DFLASH_KEY} ..."
|
|
download_weight_key "${DFLASH_KEY}"
|
|
echo ""
|
|
else
|
|
echo "[dflash] Skipping DFlash draft model. Set WITH_DFLASH_DRAFT=1 to fetch it when a matching compose requires it."
|
|
fi
|
|
|
|
echo ""
|
|
|
|
if [[ "${WITH_PRISM_EAGLE3:-0}" == "1" ]] && [[ "${SKIP_MODEL:-0}" != "1" ]]; then
|
|
case "${MODEL_NAME}" in
|
|
qwen3.6-27b)
|
|
download_weight_key qwen3.6-27b:prism_eagle3
|
|
;;
|
|
*)
|
|
echo "ERROR: WITH_PRISM_EAGLE3=1 is only wired for qwen3.6-27b." >&2
|
|
exit 1 ;;
|
|
esac
|
|
echo ""
|
|
fi
|
|
|
|
# Note: vllm#40361 Marlin pad-sub-tile-n patched files are vendored in-repo
|
|
# at models/qwen3.6-27b/vllm/patches/vllm-marlin-pad/. Dual-card composes
|
|
# mount them via repo-relative paths — no host filesystem dependency, no
|
|
# clone needed. (Previous design required cloning a fork to /opt/ai/engines/vllm/primary/;
|
|
# refactored 2026-05-03 to vendor the two files in-repo, fixing #37.)
|
|
|
|
# Per-model "next steps" — different composes / served-model-name / port between models.
|
|
SETUP_MODEL_DISPLAY="$(model_label "${MODEL_NAME}")"
|
|
case "${MODEL_NAME}" in
|
|
qwen3.6-27b)
|
|
SAMPLE_CONTAINER="vllm-qwen36-27b"
|
|
SAMPLE_COMPOSE_FLAGS_DUAL=" -f dual/autoround-int4/fp8-mtp.yml"
|
|
SAMPLE_PORT="8020"
|
|
SAMPLE_MODEL_NAME="qwen3.6-27b-autoround"
|
|
NEXT_STEPS_NOTE="Or dual-card vLLM (Marlin patched files already vendored in-repo):
|
|
cd models/${MODEL_NAME}/vllm/compose && docker compose -f dual/autoround-int4/fp8-mtp.yml up -d"
|
|
;;
|
|
gemma-4-31b)
|
|
SAMPLE_CONTAINER="vllm-gemma-4-31b-mtp"
|
|
# Gemma 4 ships specific composes — pick MTP as the canonical default.
|
|
# Use scripts/switch.sh which auto-selects the right compose by variant.
|
|
SAMPLE_COMPOSE_FLAGS_DUAL=""
|
|
SAMPLE_PORT="8030"
|
|
SAMPLE_MODEL_NAME="gemma-4-31b-autoround"
|
|
NEXT_STEPS_NOTE="Available variants:
|
|
bash scripts/switch.sh vllm/gemma-bf16-mtp # MTP drafter, TP=2, port 8030 (dual-card)
|
|
bash scripts/switch.sh beellama/gemma-dflash # DFlash, single-card default, port 8061"
|
|
;;
|
|
gemma-4-26b-a4b)
|
|
SAMPLE_CONTAINER="vllm-gemma-4-26b-a4b"
|
|
SAMPLE_COMPOSE_FLAGS_DUAL=""
|
|
SAMPLE_PORT="8035"
|
|
SAMPLE_MODEL_NAME="gemma-4-26b-a4b"
|
|
NEXT_STEPS_NOTE="Available variants:
|
|
bash scripts/switch.sh vllm/gemma-26b-awq
|
|
WITH_ASSISTANT_DRAFT=1 bash scripts/setup.sh gemma-4-26b-a4b # fetch MTP assistant if using awq-mtp"
|
|
;;
|
|
diffusiongemma-26b-a4b)
|
|
SAMPLE_CONTAINER="vllm-diffusiongemma-26b-a4b-fp8-tp2"
|
|
SAMPLE_COMPOSE_FLAGS_DUAL=""
|
|
SAMPLE_PORT="8042"
|
|
SAMPLE_MODEL_NAME="diffusiongemma-26b-a4b"
|
|
NEXT_STEPS_NOTE="🧪 experimental dLLM (vLLM's first), dual-card. Launch needs --force (non-functional status):
|
|
bash scripts/switch.sh --force vllm/diffusiongemma-dual
|
|
# or: gpu-mode dgemma (stops other GPU models, serves on :8199)"
|
|
;;
|
|
qwen3.6-35b-a3b)
|
|
SAMPLE_CONTAINER="vllm-qwen36-35b-a3b"
|
|
SAMPLE_COMPOSE_FLAGS_DUAL=""
|
|
SAMPLE_PORT="8040"
|
|
SAMPLE_MODEL_NAME="qwen3.6-35b-a3b-autoround"
|
|
NEXT_STEPS_NOTE="Preview variants:
|
|
bash scripts/switch.sh vllm/qwen35-preview"
|
|
;;
|
|
esac
|
|
|
|
echo "[setup] ✓ ${SETUP_MODEL_DISPLAY} downloaded."
|
|
echo "[setup] Next: bash scripts/launch.sh"
|
|
echo ""
|
|
echo "Next — single-card vLLM (default):"
|
|
if [[ "${MODEL_NAME}" == "gemma-4-31b" ]]; then
|
|
echo " bash scripts/switch.sh vllm/gemma-bf16-mtp"
|
|
echo " docker logs -f ${SAMPLE_CONTAINER}"
|
|
else
|
|
echo " cd models/${MODEL_NAME}/vllm/compose && docker compose up -d"
|
|
echo " docker logs -f ${SAMPLE_CONTAINER}"
|
|
fi
|
|
echo ""
|
|
echo "${NEXT_STEPS_NOTE}"
|
|
echo ""
|
|
echo "Sanity test (after 'Application startup complete'):"
|
|
echo " curl -sf http://localhost:${SAMPLE_PORT}/v1/chat/completions \\"
|
|
echo " -H 'Content-Type: application/json' \\"
|
|
echo " -d '{\"model\":\"${SAMPLE_MODEL_NAME}\",\"messages\":[{\"role\":\"user\",\"content\":\"Capital of France?\"}],\"max_tokens\":200}'"
|