Skip to content
#

multi-model-evaluation

Here are 2 public repositories matching this topic...

Single-case benchmark: AI agents making timely, auditable decisions over a continuously updating financial data stream. Alphabet Q1 2026 earnings night, real SEC filings, SQLite publication-time gating, deterministic + LLM judges. Calibrated on 5 models: kimi-k3.1-pro 58.8, glm-5.3 48.7, deepseek-flash 46.8, qwen3.8-max 44.6, deepseek-v4-pro 32.5.

  • Updated Nov 12, 2026
  • Python

Add this topic to your repo

To associate your repository with the multi-model-evaluation topic, visit your repo's landing page and select "manage topics."

Learn more