Deploy the physics-agent-service locally using Docker Compose with the bundled OVRTX GPU rendering sidecar. Use when user wants to run physics agent with docker, docker compose, set up local deployment of the physics service, run it on a GPU box, start physics agent containers, configure the VLM provider for physics docker deployment, or check whether tune/refine service routes are usable. Trigger phrases include "deploy physics agent", "docker compose physics", "run physics agent locally", "start physics service docker", "physics compose up", "physics agent docker".
Instrucciones de origen · Vista previa de solo lectura
name
deploy-physics-agent-docker
description
Deploy the physics-agent-service locally using Docker Compose with the bundled OVRTX GPU rendering sidecar. Use when user wants to run physics agent with docker, docker compose, set up local deployment of the physics service, run it on a GPU box, start physics agent containers, configure the VLM provider for physics docker deployment, or check whether tune/refine service routes are usable. Trigger phrases include "deploy physics agent", "docker compose physics", "run physics agent locally", "start physics service docker", "physics compose up", "physics agent docker".
Requires Docker daemon, Docker Compose v2.24+, NVIDIA Container Toolkit, an NVIDIA GPU with about 16GB+ VRAM for the OVRTX sidecar, repo-root .env provider credentials, free host ports 8000/8001, and a tuning-enabled image when validating service-side refine.
Deploy Physics Agent Service with Docker Compose
Deploy the physics-agent-service and the bundled OVRTX rendering API locally using Docker Compose. The physics service is CPU-only; the rendering sidecar uses the GPU.
When to Use
Use when the user wants to run physics-agent-service locally with Docker Compose.
Use when the user needs the bundled OVRTX rendering sidecar for physics classification.
Use when the user wants to configure VLM provider credentials or run local smoke requests with optimizer flags.
Use when the user asks whether /tune or /refine is available in a
Docker deployment, including the server-configured refine provider requirements.
Use quickstart for a shorter first local POC, and use deploy-collection when running multiple Content Agents together.
Limitations
The default stack owns host ports 8000 and 8001. Stop overlapping Material, Physics, Texture, or standalone OVRTX stacks before startup.
The main service waits on OVRTX readiness; OVRTX is ready only when /health reports gpu_initialized: true.
First build and first render are long-running operations. Return logs and health commands rather than holding an agent session open indefinitely.
Keep secrets out of chat and commits. Tell the user to edit .env; do not ask them to paste keys.
Service /refine requires an image with the tuning dependencies, an
OvPhysX runtime, and a registered chat/VLM provider selected with
PA_REFINE_BACKEND and PA_REFINE_MODEL.
Prerequisites
Check before deploying:
Docker Compose v2.24+: docker compose version -- required for env_file: required: false long-form syntax
NVIDIA GPU with ~16 GB+ VRAM: nvidia-smi
NVIDIA Container Toolkit installed: docker run --rm --gpus all nvidia/cuda:12.0-base nvidia-smi
VLM provider API key (at least one): NVIDIA NIM, OpenAI, Anthropic, or Gemini
Scene Optimizer backend: either run ./scripts/fetch_build_resources.sh
for the local optimizer bundle, or configure a remote NVCF optimizer with
NGC_API_KEY plus /
NVCF_OPTIMIZER_FUNCTION_ID
OPTIMIZER_ENDPOINT
Refine runtime/config only when validating /refine: tuning
dependencies, an OvPhysX runtime, PA_REFINE_BACKEND, PA_REFINE_MODEL,
and the selected provider's credential
Instructions
Confirm Docker, Compose, GPU, NVIDIA Container Toolkit, and port availability before starting the stack.
Create or update the repo-root .env with exactly the VLM provider credentials the selected backend needs.
Prepare one Scene Optimizer backend before building the image: run
./scripts/fetch_build_resources.sh for the local bundle, or skip that
local fetch and set NGC_API_KEY plus NVCF_OPTIMIZER_FUNCTION_ID /
OPTIMIZER_ENDPOINT for a remote NVCF optimizer.
Start the Physics Agent compose stack from the repo root.
Wait for both the main service and OVRTX readiness checks before reporting the service ready.
For optimizer-sensitive smoke assets, use the optimizer form fields below.
For tune/refine validation, confirm the container image includes the tuning
extra. For /refine, also confirm PA_REFINE_BACKEND, PA_REFINE_MODEL,
and the selected provider's credential before submitting a job.
Return service URLs, health state, log commands, and stop commands using the output format below.
Set VLM API Key
Create .env at the repo root (the compose file reads it via env_file: ../../.env):
Choose one Scene Optimizer backend before starting the stack.
Local Scene Optimizer Core
./scripts/fetch_build_resources.sh
docker compose -f apps/physics_agent_service/docker-compose.yml up --build
scripts/fetch_build_resources.sh stages the local Scene Optimizer Core bundle
under .build-resources/scene_optimizer_core so the default optimize_usd
pipeline path works inside the service image. If the default package is not
usable for the host architecture, set SO_CORE_URL to an explicit Scene
Optimizer Core zip.
Remote NVCF Scene Optimizer
Skip ./scripts/fetch_build_resources.sh for remote-only deployments,
especially on host architectures without a local Scene Optimizer package.
cat >> .env <<'EOF'
NGC_API_KEY=ngc-...
NVCF_OPTIMIZER_FUNCTION_ID=...
# or OPTIMIZER_ENDPOINT=https://...
EOF
docker compose -f apps/physics_agent_service/docker-compose.yml up --build
This starts:
physics-agent-service on port 8000 (REST API)
ovrtx-rendering-api on port 8001 (GPU rendering, built from source)
First build takes ~10 minutes. First render takes ~5 minutes (shader compilation; cached after).
The main service depends_on the rendering API's health check passing (which flips gpu_initialized to true). On cold start expect the physics-agent container to sit in "waiting" state for ~5 minutes before it comes up.
Operations
View Logs
# All services
docker compose -f apps/physics_agent_service/docker-compose.yml logs -f
# Specific service
docker logs physics-agent-service
docker logs physics-ovrtx-rendering-api
Stop
# Stop all services
docker compose -f apps/physics_agent_service/docker-compose.yml down
# Stop and remove session data
docker compose -f apps/physics_agent_service/docker-compose.yml down -v
Rebuild After Code Changes
docker compose -f apps/physics_agent_service/docker-compose.yml up --build
# Force full rebuild (no cache)
docker compose -f apps/physics_agent_service/docker-compose.yml build --no-cache
docker compose -f apps/physics_agent_service/docker-compose.yml up
Check Health
curl http://localhost:8000/health # main service
python - <<'PY'
import json
from urllib.request import urlopen
try:
with urlopen("http://localhost:8001/health", timeout=10) as response:
health = json.load(response)
except Exception as exc:
print(f"rendering API unreachable: {exc}")
raise SystemExit(1)
print(json.dumps(health))
if health.get("status") == "unhealthy":
print("rendering API unhealthy")
raise SystemExit(1)
if health.get("gpu_initialized") is True:
print("rendering API ready")
else:
print("rendering API warming")
PY
REST Smoke with Optimizer Flags
For ordinary smoke assets, run the pipeline without optimizer flags. For
instanced USDs, assets that fail apply_physics with an instance-proxy authoring
error, or one combined mesh that needs split-by-component predictions, pass the
optimizer form fields to POST /pipeline:
# Instance-proxy authoring fix.
curl -X POST "http://localhost:8000/pipeline" \
-F "usd_file=@scene.usd" \
-F "optimize_usd=true" \
-F "enable_deinstance=true"# Also split one combined disjoint mesh into separate component predictions.
curl -X POST "http://localhost:8000/pipeline" \
-F "usd_file=@scene.usd" \
-F "optimize_usd=true" \
-F "enable_deinstance=true" \
-F "enable_split=true"
Use enable_deduplicate=true only when repeated identical geometry should be
collapsed. At least one optimizer operation must be enabled when
optimize_usd=true.
Refine Route Smoke
Only run this on a tuning-enabled image. Production /refine execution
requires the tuning extra, an OvPhysX daemon environment, and a registered
chat/VLM provider with matching credentials.
# First produce an apply_physics output USD with /pipeline.
PIPELINE_SESSION=$(curl -fsS -X POST "http://localhost:8000/pipeline" \
-F "usd_file=@scene.usd" | jq -r .session_id)
# Wait for /pipeline/$PIPELINE_SESSION/status to become completed, then refine.
REFINE_SESSION=$(curl -fsS -X POST "http://localhost:8000/refine" \
-F "source_session_id=$PIPELINE_SESSION" \
-F "scenario_yaml=<apps/physics_agent/configs/tuning/drop_settle.yaml" \
-F "user_prompt=make the object settle on the target surface" \
-F "optimizer=botorch" \
-F "score_threshold=0.9" \
-F "seed=42" | jq -r .session_id)
curl -fsS "http://localhost:8000/refine/$REFINE_SESSION/status" | jq .
curl -fsS "http://localhost:8000/refine/$REFINE_SESSION/results" | jq .
Resource Requirements
Configuration
GPUs
CPU
Memory
Default (main + rendering)
1
10
20 G
Environment Variables
Configurable via .env at the repo root. Key settings:
Variable
Default
Description
NVIDIA_API_KEY
NVIDIA (build.nvidia.com) VLM provider
OPENAI_API_KEY
OpenAI VLM provider
ANTHROPIC_API_KEY
Anthropic VLM provider
GOOGLE_API_KEY
Google Gemini VLM provider
PA_VLM_BACKEND
nim
Which VLM backend to use
PA_VLM_MODEL
google/gemma-4-31b-it
Model id for the selected backend
PA_VLM_TEMPERATURE
1.0
Sampling temperature
PA_REFINE_BACKEND
gemini
/refine judge/refiner backend; must be registered for both chat and VLM
PA_REFINE_MODEL
gemini-3-pro-preview
/refine judge/refiner model override
PA_MAX_ACTIVE_SESSIONS
1
Max concurrent pipelines
PA_SESSION_TTL_HOURS
24
Session expiry time
PA_MAX_UPLOAD_SIZE_MB
500
Max USD upload size
WU_NVCF_GLOBAL_MAX_CONCURRENT_REQUESTS
1
Process-wide outbound render request cap for the local OVRTX sidecar
OVRTX_NUM_SENSOR_UPDATES
500
Sensor update count before capture (rendering sidecar)
GPU Configuration
To assign specific GPUs to the rendering API, edit
apps/physics_agent_service/docker-compose.yml:
ovrtx-rendering-api:deploy:resources:reservations:devices:-driver:nvidiacount:1# number of GPUscapabilities: [gpu]
Or pin a specific GPU ID:
device_ids: ['0']
Output Format
When handing control back to the user, report:
SERVICE_URL: http://localhost:8000
DOCS_URL: http://localhost:8000/docs
SERVICE_HEALTH: healthy, starting, or unhealthy
OVRTX_HEALTH: healthy only when /health contains "gpu_initialized":true; otherwise warming or unhealthy
STOP: docker compose -f apps/physics_agent_service/docker-compose.yml down
Any missing credentials, port conflicts, GPU/toolkit blockers, or optimizer flags used for smoke validation.
For /refine, whether the image has tuning/OvPhysX dependencies and whether
the selected PA_REFINE_BACKEND, PA_REFINE_MODEL, and credential are
configured.
Shader compilation on first boot takes ~5 minutes; wait it out.
Main service unhealthy before rendering API ready
The main service depends_on the rendering API's health check. If rendering takes long to start, the physics-agent-service container will stay in "waiting" state. Check docker compose ps to see which container is blocking.
503 / VLM failures under load
PA_MAX_ACTIVE_SESSIONS defaults to 1 because rendering plus a VLM call per prim is the main throughput bottleneck. Raising this requires headroom on both CPU memory and VLM provider quota.
Refine fails before first iteration
The service /refine route builds judge/refiner models inside the deployment.
If logs mention missing dependencies, backend registration, or an API key, use
an image with the tuning extra and OvPhysX runtime, then select a provider
registered for both chat and VLM and configure its credential.
