teahubxyzandClaude Fable 5.1
5f61220fdc
cloudflare/inference.js: one Worker (time-inference) at inference.<TIME_VENTURE_DOMAIN> holding VENICE_API_KEY as a secret; venture pages send the same bodies they would send Venice, with an x-app header. Allowlisted routes (chat, music quote/queue/retrieve/complete, image generate), allowlisted models, size caps, and per-person / per-app daily quotas in D1 (the same counter pattern as worker.js). inference.py deploys it; idempotent. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01K6897PWcUEwHiqZMcVnNF6
87 lines
3.3 KiB
Python
87 lines
3.3 KiB
Python
"""The shared inference Worker: Venice behind a key the pages never see (cloudflare/inference.js).
|
|
|
|
python3 .github/time/inference.py deploy # script time-inference at inference.<TIME_VENTURE_DOMAIN>
|
|
python3 .github/time/inference.py check # /health of the deployed worker
|
|
|
|
Needs CLOUDFLARE_API_TOKEN, CLOUDFLARE_ACCOUNT_ID, TIME_VENTURE_DOMAIN and VENICE_API_KEY (bound as a Worker secret).
|
|
Settings (all optional, TIME_INFERENCE_*): APPS, ORIGINS, CHAT_MODELS, AUDIO_MODELS, IMAGE_MODELS, AUDIO_SECONDS,
|
|
MAX_TOKENS, CHAT_DAILY, CHAT_TOTAL, RENDER_DAILY, RENDER_TOTAL. Idempotent: deploying again updates in place.
|
|
"""
|
|
from __future__ import annotations
|
|
|
|
import hashlib
|
|
import json
|
|
import os
|
|
import sys
|
|
import urllib.request
|
|
from pathlib import Path
|
|
|
|
sys.path.insert(0, str(Path(__file__).resolve().parent))
|
|
from cloudflare import COMPAT, account, attach, cf, env, multipart # noqa: E402
|
|
|
|
HERE = Path(__file__).resolve().parent
|
|
WORKER = HERE / "cloudflare" / "inference.js"
|
|
NAME = "time-inference"
|
|
DEFAULTS = {
|
|
"APPS": "gandora,havre",
|
|
"ORIGINS": "*.yolopreneur.lol",
|
|
"CHAT_MODELS": "claude-haiku-5-5,google-gemma-4-26b-a4b-it,mistral-small-3-2-24b-instruct,deepseek-v4-flash,qwen3-235b-a22b-instruct-2507",
|
|
"AUDIO_MODELS": "ace-step-15",
|
|
"IMAGE_MODELS": "z-image-turbo",
|
|
"AUDIO_SECONDS": "60",
|
|
"MAX_TOKENS": "2000",
|
|
"CHAT_DAILY": "60", "CHAT_TOTAL": "2000",
|
|
"RENDER_DAILY": "20", "RENDER_TOTAL": "400",
|
|
}
|
|
|
|
|
|
def settings() -> dict:
|
|
return {k: env("TIME_INFERENCE_" + k, v) for k, v in DEFAULTS.items()}
|
|
|
|
|
|
def hostname() -> str:
|
|
zone = env("TIME_VENTURE_DOMAIN").lower().strip(".")
|
|
if not zone:
|
|
raise SystemExit("set TIME_VENTURE_DOMAIN (the zone the worker lives under)")
|
|
return f"inference.{zone}"
|
|
|
|
|
|
def database() -> str:
|
|
for d in cf("GET", f"{account()}/d1/database?name={NAME}") or []:
|
|
if d.get("name") == NAME:
|
|
return d["uuid"]
|
|
return cf("POST", f"{account()}/d1/database", {"name": NAME})["uuid"]
|
|
|
|
|
|
def deploy() -> dict:
|
|
key = env("VENICE_API_KEY")
|
|
if not key:
|
|
raise SystemExit("set VENICE_API_KEY: it becomes the worker's secret")
|
|
vars_ = settings()
|
|
h = hashlib.sha256(WORKER.read_bytes())
|
|
h.update(json.dumps({**vars_, "host": hostname(), "compat": COMPAT}, sort_keys=True).encode())
|
|
vars_["REV"] = rev = h.hexdigest()[:12]
|
|
db = database()
|
|
meta = {"main_module": "inference.js", "compatibility_date": COMPAT,
|
|
"bindings": [{"type": "d1", "name": "DB", "id": db}, {"type": "secret_text", "name": "VENICE_API_KEY", "text": key}]
|
|
+ [{"type": "plain_text", "name": k, "text": str(v)} for k, v in vars_.items()]}
|
|
data, ctype = multipart([("metadata", "application/json", json.dumps(meta).encode()),
|
|
("inference.js", "application/javascript+module", WORKER.read_bytes())])
|
|
cf("PUT", f"{account()}/workers/scripts/{NAME}", data=data, ctype=ctype)
|
|
url = attach(NAME, hostname())
|
|
return {"url": url, "worker": NAME, "db": db, "rev": rev}
|
|
|
|
|
|
def check() -> dict:
|
|
with urllib.request.urlopen(f"https://{hostname()}/health", timeout=20) as r:
|
|
return json.loads(r.read())
|
|
|
|
|
|
if __name__ == "__main__":
|
|
if sys.argv[1:2] == ["deploy"]:
|
|
print(json.dumps(deploy(), indent=1))
|
|
elif sys.argv[1:2] == ["check"]:
|
|
print(json.dumps(check(), indent=1))
|
|
else:
|
|
sys.exit(__doc__)
|