🧪 Local LLM Benchmark Suite

Grades local LLM models (run via LM Studio on an Apple M3 Max / 48 GB) on a strict systems-coding prompt — a pure-stdlib Python concurrent async LFU cache with TTL eviction and ACID transactions — and renders the results into a cyberpunk-terminal dashboard.

What's in here

prompts/
  lfu_cache_prompt.txt   # the exam prompt every model gets
  grading.txt            # the grader's rubric (5 pillars × 20 pts = 100)
outputs/                 # raw model .py outputs (named <model>-<quant>.py)
data/
  benchmark_history.json # persistent results store (source of truth)
generate_dashboard.py    # reads the JSON → builds dashboard.html + pages/*.html
dashboard.html           # generated — main leaderboard + charts  (gitignored)
pages/                   # generated — per-model detail pages        (gitignored)

Workflow

1. Feed prompts/lfu_cache_prompt.txt to a model in LM Studio.

2. Save its output to outputs/<model>-<quant>.py.

3. Grade it (audit the 5 pillars, capture tok/sec + tokens + TTFT), and append its entry to data/benchmark_history.json. See prompts/grading.txt for the rubric.

4. Regenerate the site:

   python3 generate_dashboard.py
   

   This (re)writes dashboard.html and every pages/<model>.html.

Viewing locally

python3 -m http.server 8000
# open http://localhost:8000/dashboard.html

Deploying (Gitea + Coolify)

The repo holds source only (outputs/, data/, generate_dashboard.py, prompts/). The generated dashboard.html and pages/ are gitignored — Coolify runs python3 generate_dashboard.py as a build step, then serves the static files. See DEPLOY.md for the exact Coolify service config.

Hardware

Apple M3 Max, 48 GB unified memory. Local inference via LM Studio. A cloud model (DeepSeek V4 Flash) is included as a quality baseline.