Claude
084922afaa
Bring the full EditmaskwithAI application into the repo under paintplus/ (429 files) so the service is self-contained — the installer copies the vendored source to ~/docker/paintplus/src instead of cloning at runtime. Rename to PaintPlus (service + branding; app logic untouched): - services/editmaskwithai.sh -> services/paintplus.sh (register_service paintplus, install_paintplus, ~/docker/paintplus, Caddy paintplus:8000, Authelia option preserved) - container names -> paintplus across docker-compose*.yml; dev network -> paintplus-network - browser <title> -> "PaintPlus - AI Image Editor"; README heading -> PaintPlus with upstream provenance note - README utilities table: editmaskwithai -> paintplus Backend/frontend code (help strings referencing the old container name, the ai_photo_edit.db filename) is intentionally left as-is to avoid touching application logic. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Nb2vJ8W7bHKx1JXVvpCraH
97 lines
3.1 KiB
Python
97 lines
3.1 KiB
Python
"""
|
|
GPU status and model management endpoints.
|
|
All under /api/gpu prefix.
|
|
"""
|
|
from fastapi import APIRouter
|
|
from pydantic import BaseModel
|
|
from typing import Optional, List
|
|
import asyncio
|
|
|
|
router = APIRouter(prefix="/api/gpu", tags=["gpu"])
|
|
|
|
|
|
@router.get("/status")
|
|
async def gpu_status():
|
|
"""
|
|
Full GPU capability report: hardware, feature flags, VRAM budget,
|
|
and which model was selected for each operation.
|
|
Frontend polls this to show GPU badge and tool availability.
|
|
"""
|
|
from app.services.gpu_detect import get_cached_gpu_info
|
|
from app.services.local_diffusion import get_all_model_states
|
|
|
|
info = get_cached_gpu_info()
|
|
|
|
return {
|
|
# Hardware
|
|
"backend": info.backend,
|
|
"device_name": info.device_name,
|
|
"vram_total_gb": info.vram_total_gb,
|
|
"vram_free_gb": info.vram_free_gb,
|
|
"compute_capability": info.compute_capability,
|
|
# Feature flags
|
|
"fp16": info.fp16,
|
|
"bf16": info.bf16,
|
|
"fp8": info.fp8,
|
|
"int8": info.int8,
|
|
"tensor_cores": info.tensor_cores,
|
|
"xformers": info.xformers,
|
|
# Derived
|
|
"effective_vram_gb": info.effective_vram_gb,
|
|
"tier": info.tier,
|
|
# Selected models per operation
|
|
"recommended": {
|
|
op: (
|
|
{
|
|
"model_id": spec.model_id,
|
|
"family": spec.family,
|
|
"memory_opt": spec.memory_opt,
|
|
"native_res": spec.native_res,
|
|
"vram_fp16_gb": spec.vram_fp16_gb,
|
|
}
|
|
if spec else None
|
|
)
|
|
for op, spec in info.recommended.items()
|
|
},
|
|
"pipeline_states": get_all_model_states(),
|
|
"warnings": info.warnings,
|
|
"capabilities": info.capabilities,
|
|
}
|
|
|
|
|
|
class PrefetchRequest(BaseModel):
|
|
operations: Optional[List[str]] = None
|
|
|
|
|
|
@router.post("/prefetch")
|
|
async def prefetch_models(req: PrefetchRequest = PrefetchRequest()):
|
|
"""
|
|
Eagerly load pipelines into GPU memory for the requested operations.
|
|
Returns immediately; poll /api/gpu/prefetch-status for progress.
|
|
Default: inpaint, txt2img, img2img.
|
|
"""
|
|
ops = req.operations or ["inpaint", "txt2img", "img2img"]
|
|
valid = {"inpaint", "txt2img", "img2img", "outpaint", "upscale"}
|
|
ops = [op for op in ops if op in valid]
|
|
|
|
from app.services.local_diffusion import get_local_diffusion_provider
|
|
provider = get_local_diffusion_provider()
|
|
|
|
async def _prefetch():
|
|
for op in ops:
|
|
try:
|
|
await provider._get_pipeline(op)
|
|
print(f"[gpu] Prefetch complete: {op}")
|
|
except Exception as exc:
|
|
print(f"[gpu] Prefetch failed for {op}: {exc}")
|
|
|
|
asyncio.create_task(_prefetch())
|
|
return {"status": "prefetch_started", "operations": ops}
|
|
|
|
|
|
@router.get("/prefetch-status")
|
|
async def prefetch_status():
|
|
"""Poll model download / load progress."""
|
|
from app.services.local_diffusion import get_all_model_states
|
|
return {"models": get_all_model_states()}
|