Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
A benchmark for evaluating LLMs on Chinese traditional fortune telling — Bazi (八字) and Ziwei Doushu (紫微斗数).
| Date | Stars |
|---|---|
| 2026-07-31 | 2236 |
| 2026-08-04 | 2250 |
| 2026-08-06 | 2250 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
<div align="center"> # Chinese Fortune Telling Bench **A benchmark for evaluating large language models on Chinese traditional fortune telling — Bazi (八字) and Ziwei Doushu (紫微斗数).** [](./LICENSE) [](https://www.python.org/) [](./data/data.json) [](./data) English | [中文](./README_zh.md) </div> --- ## Overview Questions are multiple-choice and scored by exact match against a ground-truth answer. The corpus is sourced from the annual **Global Fortune Teller Competition** ([全球算命师大赛](https://hkjfma.org)) for 2022–2025. Raw sheets live under [data/raw/](./data/raw/). For Bazi-specific benchmarks, [BaziQA (Chen et al., 2025)](https://arxiv.org/pdf/2602.12889). | File | Description | |---|---| | [data/data.json](./data/data.json) | 160 normalized multiple-choice questions across twelve life aspects (career, health, marriage, children, wealth, …). | | [data/fortune_api_results.json](./data/fortune_api_results.json) | Pre-computed Bazi and Ziwei charts (via [iztro](https://github.com/SylarLong/iztro)), keyed by `case_id` and joined to `data.json` at runtime when `--astro` is set. Isolates pure reasoning from chart derivation. | > Filter by year with `--year`. Inspect the dataset with `--stats`. --- ## Installation ```bash git clone https://github.com/DestinyLinker/MingLi-Bench.git cd MingLi-Bench pip install -r requirements.txt ``` --- ## Configuration The CLI reads API keys and defaults from a `.env` file. Copy the template and fill in only the providers you plan to use — empty or placeholder values are skipped automatically. ```bash cp .env.example .env ``` <details> <summary><b>Supported providers</b></summary> ```dotenv # OpenRouter — one key, most models OPENROUTER_API_KEY=sk-or-... OPENROUTER_BASE_URL=https://openrouter.ai/api/v1 # Native providers (only if you call them directly) OPENAI_API_KEY=sk-... # OPENAI_BASE_URL=https://api.openai.com/v1 # override for OpenAI-compatible gateways ANTHROPIC_API_KEY=sk-ant-... GOOGLE_API_KEY=... DEEPSEEK_API_KEY=sk-... DEEPSEEK_BASE_URL=https://api.deepseek.com # Doubao / Volcengine — both key and endpoint id are required DOUBAO_API_KEY=... DOUBAO_BASE_URL=https://ark.cn-beijing.volces.com/api/v3 DOUBAO_ENDPOINT_ID=ep-... # Defaults (CLI flags can override) TIMEOUT=60 MAX_WORKERS=5 MAX_TOKENS=8192 TEMPERATURE=0.0 ``` </details> `.env` is git-ignored. To run against a different key set, pass `--env-file /path/to/other.env`. **Sanity check** once keys are in place: ```bash python -m mingli_bench.cli --list-models # supported model names python -m mingli_bench.cli --stats # dataset statistics ``` --- ## Quick Start > **Recommended defaults:** always pass `--cot` and `--astro`. Chain-of-Thought gives the model room to reason through the chart, and `--astro` injects pre-computed Bazi/Ziwei charts so scores reflect reasoning ability rather than date-to-chart conversion accuracy. Drop them only when you specifically want to ablate either effect. There are two common routing modes. ### 1. Via OpenRouter (one key, most models) Leave `--platform` unset and write the model name in OpenRouter's `provider/model` form. The CLI auto-detects the `/` and routes through OpenRouter. ```bash python -m mingli_bench.cli --model openai/gpt-4o --year 2025 --cot --astro --max-workers 8 python -m mingli_bench.cli --model anthropic/claude-sonnet-4-6 --year 2025 --cot --astro python -m mingli_bench.cli --model google/gemini-2.5-pro --year 2025 --cot --astro python -m mingli_bench.cli --model deepseek/deepseek-r1 --year 2025 --cot --astro ``` ### 2. Native provider When the model name doesn't match auto-detection rules (e.g. a versioned Doubao endpoint id), or when you want to force an OpenA
Excerpt of 6,623 characters
Read on GitHub
8
Would you bet a product on this? Bounded 0–100 and slow moving.
matched fp:61a3612ecbd3d56f, llm:repository description: 'A benchmark for evaluating LLMs on Chinese traditional fortune telling — Bazi (八字) and Ziwei Doushu (紫微斗数).'
matched fp:61a3612ecbd3d56f, llm:repository description: 'A benchmark for evaluating LLMs on Chinese traditional fortune telling — Bazi (八字) and Ziwei Doushu (紫微斗数).'