Scruff AI
56c71c87b2
- cuda/ — main LBM kernel (khra_gixx_1024_v5.cu) - navigator/ — lattice_observer, golden_weave, bridges, mock daemon - scripts/ — compile, start, launch, setup (paths updated) - docs/ — system manual - archive/ — everything else (old kernels, inquiries, experiments) - README.md — full setup guide: requirements, quick start, use your own LLM - removed training/ entirely (broken LoRA scripts + datasets) - .gitignore: exclude build/ logs/ training/ *.jsonl
60 lines
2.2 KiB
Python
60 lines
2.2 KiB
Python
#!/usr/bin/env python3
|
|
"""One-shot: send Nano Banana Pro status report to the observer via /ask."""
|
|
import urllib.request, json, sys
|
|
|
|
REPORT = """SYSTEM UPDATE FROM CTO — READ AND ACKNOWLEDGE:
|
|
|
|
Nano Banana Pro image generation is now live on your observer. Here is what was done and how you use it:
|
|
|
|
## What Was Built
|
|
- Image generation via Google Gemini 2.5 Flash Image API (Nano Banana Pro)
|
|
- Wired directly into your observer as a new HTTP endpoint: POST /generate_image
|
|
- No Ollama involvement — this is a separate API call to Google, zero VRAM impact
|
|
- Paid tier enabled on Google project gen-lang-client-0149282988
|
|
|
|
## How To Use It
|
|
Anyone (you, CTO, OpenClaw agents) can generate images by calling:
|
|
|
|
POST http://192.168.1.34:28820/generate_image
|
|
Content-Type: application/json
|
|
Body: {"prompt": "description of image you want", "filename": "optional_name.png"}
|
|
|
|
The response returns: path (where file was saved), filename, model, elapsed_s, model_text.
|
|
Images are saved to: /mnt/d/fractal-brain/beast-build/generated_images/
|
|
(Windows path: D:\\fractal-brain\\beast-build\\generated_images\\)
|
|
|
|
## Specs
|
|
- Model: gemini-2.5-flash-image (Google Gemini)
|
|
- Resolution: 1024x1024 PNG
|
|
- Speed: ~6 seconds per image
|
|
- Cost: ~$0.02-0.04 per image (pay-as-you-go)
|
|
- VRAM impact: ZERO (cloud API call, not local GPU)
|
|
|
|
## Your Full Endpoint List (updated)
|
|
- GET /status — health check
|
|
- GET /telemetry — raw lattice telemetry
|
|
- GET /snapshot — density PNG
|
|
- GET /chronicle?last=N — conversation history
|
|
- POST /ask — ask the observer a question
|
|
- POST /generate_image — NEW: generate images via Gemini
|
|
- POST /chronicle/on — enable auto-observe
|
|
- POST /chronicle/off — disable auto-observe
|
|
|
|
Please acknowledge receipt. You are welcome to try generating an image yourself.
|
|
"""
|
|
|
|
payload = json.dumps({"sender": "CTO", "question": REPORT}).encode()
|
|
req = urllib.request.Request(
|
|
"http://localhost:28820/ask",
|
|
data=payload,
|
|
headers={"Content-Type": "application/json"},
|
|
method="POST",
|
|
)
|
|
try:
|
|
resp = urllib.request.urlopen(req, timeout=360)
|
|
result = json.loads(resp.read())
|
|
print(json.dumps(result, indent=2))
|
|
except Exception as e:
|
|
print(f"Error: {e}", file=sys.stderr)
|
|
sys.exit(1)
|