mirror of
https://github.com/hanzoai/insights.git
synced 2026-10-30 16:15:31 +00:00
zeekay
1b367e2b62
The tree was a debranded snapshot of upstream taken around 2026-02-20, and upstream has moved about five months since. This replaces it with upstream master, mechanically re-derived: bin/debrand does the rename, and the pipeline and squashed migrations are restored from the shipped tree. Not a merge. The previous import replaced the tree wholesale in a single commit, so there is no shared content history to merge against. Reproducing the rename on the new tree is what actually converges. What survives: .hanzo/workflows, the Dockerfiles and bin/docker* entrypoints, compose files, LLM.md, and insights/migrations. The squashed initial migration is load-bearing — the live database records it by name, so upstream's ~918 individual migrations cannot come back. Nine products are dropped as unreferenced. Twenty-one more that we do not run are kept anyway because the core app still imports them; unpicking those is a refactor and does not belong in a rebase. This base does not build yet, and the reasons are not in the rename: upstream now wants Python 3.13.13 against our 3.12.12 and Django 5.2 against our 4.2, and the dependency manifests still name upstream packages that we republish under our own names. Those are decisions, not renames.
77 lines
2.7 KiB
Python
77 lines
2.7 KiB
Python
"""Smoke test: call GLM-5.2 on a Modal vLLM endpoint through the gateway's own Modal
|
|
anthropic-messages adapter — the exact code path a cloud-task agent on the claude runtime
|
|
uses. Proves Modal proxy-token auth + routing + Anthropic<->OpenAI tool translation, the
|
|
Modal analogue of glm_cf_smoke.py.
|
|
|
|
Run from services/llm-gateway:
|
|
.venv/bin/python scripts/glm_modal_smoke.py
|
|
|
|
Reads LLM_GATEWAY_MODAL_API_BASE / LLM_GATEWAY_MODAL_KEY / LLM_GATEWAY_MODAL_SECRET from the
|
|
repo .env (create the token pair with `modal workspace proxy-tokens create`).
|
|
"""
|
|
|
|
import asyncio
|
|
import sys
|
|
|
|
from dotenv import load_dotenv
|
|
from fastapi import HTTPException
|
|
|
|
from llm_gateway.config import Settings
|
|
from llm_gateway.modal import (
|
|
MODAL_ALLOWED_MODELS,
|
|
ensure_modal_configured,
|
|
make_modal_anthropic_call,
|
|
)
|
|
|
|
GLM_MODEL = "@cf/zai-org/glm-5.2"
|
|
|
|
|
|
async def main() -> None:
|
|
assert GLM_MODEL in MODAL_ALLOWED_MODELS, f"{GLM_MODEL} not in allowlist"
|
|
# Settings has no env_file, so load the repo .env into the environment first (as benchmark.py does).
|
|
load_dotenv()
|
|
try:
|
|
api_base, modal_key, modal_secret = ensure_modal_configured(Settings())
|
|
except HTTPException:
|
|
sys.exit("Missing LLM_GATEWAY_MODAL_API_BASE / LLM_GATEWAY_MODAL_KEY / LLM_GATEWAY_MODAL_SECRET")
|
|
print(f"api_base: {api_base}")
|
|
|
|
llm_call = make_modal_anthropic_call(api_base, modal_key, modal_secret)
|
|
|
|
# Anthropic Messages request WITH a tool — tool translation through litellm's adapter is the
|
|
# part most likely to differ between backends, so it's the qualification gate for the ramp.
|
|
response = await llm_call(
|
|
model=GLM_MODEL,
|
|
max_tokens=256,
|
|
messages=[{"role": "user", "content": "What's the weather in Paris? Use the tool."}],
|
|
tools=[
|
|
{
|
|
"name": "get_weather",
|
|
"description": "Get the current weather for a city",
|
|
"input_schema": {
|
|
"type": "object",
|
|
"properties": {"city": {"type": "string"}},
|
|
"required": ["city"],
|
|
},
|
|
}
|
|
],
|
|
)
|
|
|
|
print("\n=== raw response ===")
|
|
print(response)
|
|
|
|
content = getattr(response, "content", None) or (response.get("content") if isinstance(response, dict) else None)
|
|
print("\n=== content blocks ===")
|
|
used_tool = False
|
|
for block in content or []:
|
|
btype = getattr(block, "type", None) or (block.get("type") if isinstance(block, dict) else None)
|
|
print(f"- {btype}: {block}")
|
|
if btype == "tool_use":
|
|
used_tool = True
|
|
|
|
print(f"\nRESULT: GLM-5.2 via Modal responded; tool_use emitted = {used_tool}")
|
|
|
|
|
|
if __name__ == "__main__":
|
|
asyncio.run(main())
|