Files
revids/app/service.py
T
mteehan 2d50679ec0 Initial commit: Revids REST API for LTX-2.3 image-to-video generation
- FastAPI with job polling: POST /generate, GET /jobs/{id}, /download, /health
- SQLite job persistence (aiosqlite), background task processing
- LTX-2.3 DistilledPipeline via native ltx-pipelines (editable install from submodule)
- Configurable model paths, LoRA support, FP8 quantization via env vars
- Single-concurrency GPU lock for safe inference
- LTX-2 as git submodule under libs/
2026-05-31 06:03:08 -04:00

178 lines
6.2 KiB
Python

from __future__ import annotations
import asyncio
import os
import uuid
from datetime import datetime, timezone
from typing import Optional
import torch
from ltx_pipelines import DistilledPipeline
from PIL import Image
from app.config import settings
from app.database import JobDB, JobRecord, VALID_FRAME_COUNTS
class LtxService:
def __init__(self, db: JobDB) -> None:
self.db = db
self.pipeline: Optional[DistilledPipeline] = None
self._load_lock = asyncio.Lock()
self._job_semaphore = asyncio.Semaphore(settings.max_concurrent_jobs)
self._video_dir = settings.video_output_dir
async def load_pipeline(self) -> None:
async with self._load_lock:
if self.pipeline is not None:
return
lora_config = self._parse_lora_paths(settings.ltx_loras)
self.pipeline = await asyncio.to_thread(
DistilledPipeline,
distilled_checkpoint_path=settings.ltx_distilled_checkpoint,
gemma_root=settings.ltx_gemma_root,
spatial_upsampler_path=settings.ltx_spatial_upsampler,
loras=lora_config,
device=settings.ltx_device,
quantization=settings.ltx_quantization,
)
@staticmethod
def _parse_lora_paths(lora_specs: list[str]) -> list[dict]:
result = []
for spec in lora_specs:
if ":" in spec:
path, scale = spec.rsplit(":", 1)
result.append({"path": path.strip(), "scale": float(scale.strip())})
else:
result.append({"path": spec.strip(), "scale": 1.0})
return result
async def submit_job(self, image: Image.Image, request_params: dict) -> JobRecord:
job_id = uuid.uuid4().hex[:12]
now = datetime.now(timezone.utc).isoformat()
prompt = request_params.get("prompt")
width = request_params.get("width", settings.default_width)
height = request_params.get("height", settings.default_height)
num_frames = request_params.get("num_frames", settings.default_frames)
fps = request_params.get("fps", settings.default_fps)
seed = request_params.get("seed")
if num_frames not in VALID_FRAME_COUNTS:
candidates = sorted(
f for f in VALID_FRAME_COUNTS if abs(f - num_frames) <= 8
)
raise ValueError(
f"num_frames must be 8n+1. Got {num_frames}. "
f"Closest valid: {candidates[:4]}"
)
job = JobRecord(
id=job_id,
status="pending",
prompt=prompt,
params={
"width": width,
"height": height,
"num_frames": num_frames,
"fps": fps,
"seed": seed,
},
created_at=now,
)
await self.db.create_job(job)
asyncio.create_task(self._process_job(job_id, image, job.params))
return job
async def _process_job(
self,
job_id: str,
image: Image.Image,
params: dict,
) -> None:
async with self._job_semaphore:
try:
await self.db.update_status(job_id, "processing")
await self.load_pipeline()
prompt = params.get("prompt", "") or ""
width = params.get("width", settings.default_width)
height = params.get("height", settings.default_height)
num_frames = params.get("num_frames", settings.default_frames)
fps = params.get("fps", settings.default_fps)
seed = params.get("seed") if params.get("seed") is not None else 42
if image.mode != "RGB":
image = image.convert("RGB")
image = image.resize((width, height), Image.LANCZOS)
output = await asyncio.to_thread(
self.pipeline,
prompt=prompt,
images=image,
width=width,
height=height,
num_frames=num_frames,
frame_rate=fps,
seed=seed,
)
video_path = os.path.join(self._video_dir, f"{job_id}.mp4")
os.makedirs(self._video_dir, exist_ok=True)
await asyncio.to_thread(self._save_video, output, video_path, fps)
await self.db.update_status(job_id, "completed")
except Exception as e:
await self.db.update_status(job_id, "failed", error=str(e))
@staticmethod
def _save_video(output: any, path: str, fps: float) -> None:
import imageio.v3 as iio
if isinstance(output, torch.Tensor):
frames = output.cpu().numpy()
elif isinstance(output, dict) and "video" in output:
v = output["video"]
frames = v.cpu().numpy() if isinstance(v, torch.Tensor) else v
else:
frames = output
if isinstance(frames, torch.Tensor):
frames = frames.cpu().numpy()
if frames.ndim == 4:
frames = frames[0]
if frames.max() <= 1.0:
frames = (frames * 255).astype("uint8")
elif frames.dtype != "uint8":
frames = frames.clip(0, 255).astype("uint8")
iio.imwrite(path, frames, fps=float(fps), codec="libx264")
async def get_job(self, job_id: str) -> JobRecord | None:
return await self.db.get_job(job_id)
async def delete_job(self, job_id: str) -> tuple[bool, str | None]:
video_path = os.path.join(self._video_dir, f"{job_id}.mp4")
deleted = await self.db.delete_job(job_id)
if os.path.exists(video_path):
os.remove(video_path)
return deleted, job_id
async def get_video_path(self, job_id: str) -> str | None:
video_path = os.path.join(self._video_dir, f"{job_id}.mp4")
if os.path.exists(video_path):
return video_path
return None
_service: Optional[LtxService] = None
async def get_service() -> LtxService:
global _service
if _service is None:
db = JobDB(settings.db_path)
await db.init()
_service = LtxService(db)
return _service