Zayden-Jung/logic-skeleton-bench
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Restore portable evaluation workflow for configured runner

Browse Source
This commit is contained in:
Codex Bot
2026-09-22 22:42:34 +08:00
parent 6d91091bb2
commit 40cbae289e
5 changed files with 102 additions and 24 deletions
Download Patch File Download Diff File Expand all files Collapse all files
.env.example
+3 -3
View File
@@ -1,6 +1,6 @@
# Shared fallback. Store the real value as a Gitea Actions secret, never here.
OPENAI_API_KEY=
OPENAI_BASE_URL=https://api.openai.com/v1
# Provider-neutral shared fallback. Store real keys as Gitea Actions secrets.
LLM_API_KEY=
LLM_BASE_URL=
# Renderer and solver can use different OpenAI-compatible providers/models.
RENDERER_API_KEY=
.gitea/workflows/short-eval.yml
+42
View File
@@ -0,0 +1,42 @@
name: Short abstraction evaluation
on:
workflow_dispatch:
jobs:
short-eval:
runs-on: minecraft-and-affine-server
timeout-minutes: 15
env:
LLM_API_KEY: ${{ secrets.LLM_API_KEY }}
LLM_BASE_URL: ${{ vars.LLM_BASE_URL }}
RENDERER_API_KEY: ${{ secrets.RENDERER_API_KEY }}
RENDERER_BASE_URL: ${{ vars.RENDERER_BASE_URL }}
RENDERER_API_STYLE: ${{ vars.RENDERER_API_STYLE }}
RENDERER_MODEL: ${{ vars.RENDERER_MODEL }}
SOLVER_API_KEY: ${{ secrets.SOLVER_API_KEY }}
SOLVER_BASE_URL: ${{ vars.SOLVER_BASE_URL }}
SOLVER_API_STYLE: ${{ vars.SOLVER_API_STYLE }}
SOLVER_MODEL: ${{ vars.SOLVER_MODEL }}
steps:
- name: Check out repository
uses: actions/checkout@v4
- name: Run unit tests
run: python3 -m unittest discover -s tests -v
- name: Run three short live cases
run: python3 -m logicbench run tasks/short-smoke.json
- name: Show evaluation records and summary
if: always()
run: |
test -f artifacts/short-smoke/results.jsonl && cat artifacts/short-smoke/results.jsonl || true
test -f artifacts/short-smoke/summary.json && cat artifacts/short-smoke/summary.json || true
- name: Preserve evaluation artifacts
if: always()
uses: actions/upload-artifact@v3
with:
name: short-abstraction-evaluation
path: artifacts/short-smoke/
README.md
+11 -19
View File
@@ -31,7 +31,8 @@ python3 -m logicbench generate --count 3 --length short --seed 20260922
Run the checked-in live task:
```bash
export OPENAI_API_KEY='...'
export LLM_API_KEY='...'
export LLM_BASE_URL='your-provider-compatible-base-url'
export RENDERER_MODEL='your-renderer-model'
export SOLVER_MODEL='your-solver-model'
python3 -m logicbench run tasks/short-smoke.json
@@ -41,8 +42,10 @@ Role-specific values override the shared fallback:
| Variable | Meaning |
| --- | --- |
| `RENDERER_API_KEY`, `SOLVER_API_KEY` | Optional separate keys; each falls back to `OPENAI_API_KEY` |
| `RENDERER_BASE_URL`, `SOLVER_BASE_URL` | Optional endpoints; each falls back to `OPENAI_BASE_URL`, then `https://api.openai.com/v1` |
| `LLM_API_KEY` | Optional shared key when both roles use the same account/provider |
| `RENDERER_API_KEY`, `SOLVER_API_KEY` | Optional role-specific keys; each takes precedence over `LLM_API_KEY` |
| `LLM_BASE_URL` | Optional shared OpenAI-compatible endpoint |
| `RENDERER_BASE_URL`, `SOLVER_BASE_URL` | Optional role-specific endpoints; each takes precedence over `LLM_BASE_URL` |
| `RENDERER_API_STYLE`, `SOLVER_API_STYLE` | `responses` (default) or `chat_completions` |
| `RENDERER_MODEL`, `SOLVER_MODEL` | Required, independently selected model IDs |
@@ -60,25 +63,14 @@ The run log prints the rendered questions and scores. The same records, original
Create these repository Actions settings:
1. Secret `OPENAI_API_KEY`, or separate secrets `RENDERER_API_KEY` and `SOLVER_API_KEY`.
1. Secret `LLM_API_KEY` when both roles share credentials, or role-specific secrets `RENDERER_API_KEY` and `SOLVER_API_KEY` when they do not.
2. Variables `RENDERER_MODEL` and `SOLVER_MODEL`.
3. Optional variables `RENDERER_BASE_URL`, `SOLVER_BASE_URL`, `RENDERER_API_STYLE`, and `SOLVER_API_STYLE`. Leave base URLs empty for OpenAI; use `responses` as the API style unless a compatible provider only supports chat completions.
3. Variable `LLM_BASE_URL` for a shared provider endpoint, or role-specific variables `RENDERER_BASE_URL` and `SOLVER_BASE_URL` for different providers.
4. Optional variables `RENDERER_API_STYLE` and `SOLVER_API_STYLE`, each set to `responses` or `chat_completions` according to that provider/model.
The workflow targets `runs-on: logic-bench`. Add this label to the existing `act_runner` using one of these methods:
The workflow targets the existing `minecraft-and-affine-server` runner label. Verify that a runner advertising this label is online, then manually dispatch **Short abstraction evaluation** from the repository's Actions page.
- In the Gitea UI, open the runner settings, edit its labels, and add `logic-bench:docker://docker.gitea.com/runner-images:ubuntu-latest`.
- Or add the same string to `runner.labels` in the runner's `config.yaml`, preserving existing labels, then restart `act_runner` with that config.
- Or start the daemon with `--labels` / `GITEA_RUNNER_LABELS`; this has higher precedence than the config file and must include every label the runner should continue advertising.
Example config fragment:
```yaml
runner:
labels:
- "logic-bench:docker://docker.gitea.com/runner-images:ubuntu-latest"
```
After restarting the runner, verify that `logic-bench` appears in its Gitea settings, then manually dispatch **Short abstraction evaluation** from the repository's Actions page.
For a current all-Qwen run, create only the shared `LLM_API_KEY` secret with the Model Studio key. Set `LLM_BASE_URL` to the API Host shown for that key and region, set the two model variables independently, and use `chat_completions` for both API-style variables. For example, a fast renderer and stronger solver can use `qwen3.8-flash` and `qwen3.8-max`; both currently support JSON Schema structured output. Do not hard-code the DashScope key or endpoint in the workflow.
## Tests
logicbench/client.py
+7 -2
View File
@@ -20,9 +20,14 @@ class LLMClientConfig:
def from_env(cls, role: str) -> "LLMClientConfig":
prefix = role.upper()
model = os.getenv(f"{prefix}_MODEL", "").strip()
api_key = os.getenv(f"{prefix}_API_KEY", "").strip() or os.getenv("OPENAI_API_KEY", "").strip()
api_key = (
os.getenv(f"{prefix}_API_KEY", "").strip()
or os.getenv("LLM_API_KEY", "").strip()
or os.getenv("OPENAI_API_KEY", "").strip()
)
base_url = (
os.getenv(f"{prefix}_BASE_URL", "").strip()
or os.getenv("LLM_BASE_URL", "").strip()
or os.getenv("OPENAI_BASE_URL", "").strip()
or "https://api.openai.com/v1"
)
@@ -30,7 +35,7 @@ class LLMClientConfig:
if not model:
raise ValueError(f"{prefix}_MODEL is required")
if not api_key:
raise ValueError(f"{prefix}_API_KEY or OPENAI_API_KEY is required")
raise ValueError(f"{prefix}_API_KEY or LLM_API_KEY is required")
if api_style not in {"responses", "chat_completions"}:
raise ValueError(f"{prefix}_API_STYLE must be responses or chat_completions")
return cls(model=model, api_key=api_key, base_url=base_url.rstrip("/"), api_style=api_style)
tests/test_client.py
+39
View File
@@ -0,0 +1,39 @@
import os
import unittest
from unittest.mock import patch
from logicbench.client import LLMClientConfig
class ClientConfigTests(unittest.TestCase):
def test_shared_provider_defaults_apply_to_both_roles(self):
values = {
"LLM_API_KEY": "shared-key",
"LLM_BASE_URL": "https://shared.example/v1",
"RENDERER_MODEL": "renderer-model",
"SOLVER_MODEL": "solver-model",
}
with patch.dict(os.environ, values, clear=True):
renderer = LLMClientConfig.from_env("renderer")
solver = LLMClientConfig.from_env("solver")
self.assertEqual(renderer.api_key, "shared-key")
self.assertEqual(solver.api_key, "shared-key")
self.assertEqual(renderer.base_url, "https://shared.example/v1")
self.assertEqual(solver.base_url, "https://shared.example/v1")
def test_role_specific_values_override_shared_provider(self):
values = {
"LLM_API_KEY": "shared-key",
"LLM_BASE_URL": "https://shared.example/v1",
"RENDERER_API_KEY": "renderer-key",
"RENDERER_BASE_URL": "https://renderer.example/v1/",
"RENDERER_MODEL": "renderer-model",
}
with patch.dict(os.environ, values, clear=True):
renderer = LLMClientConfig.from_env("renderer")
self.assertEqual(renderer.api_key, "renderer-key")
self.assertEqual(renderer.base_url, "https://renderer.example/v1")
if __name__ == "__main__":
unittest.main()