llmevals

LLM evals

Explorations in LLM evaluations.

• 13 Oct 2026: How large can frontier models multiply numbers?
• 30 Sep 2026. SLM vs LLM on FinQA
• 27 Sep 2026: Qwen 3.6 vs Gemma 4 E4B vs Codex coding-agent benchmark
• 25 Sep 2026: What GPT Image 2.5 Flare quality actually buys you
• 18 Sep 2026: Jev vs frontier LLMs on BANKING77
• 9 Sep 2026: Calibrating LLM confidence scores
• 13 Oct 2025: Comparing AI coding agents on quality, cost, speed
• 27 Sep 2025: Bring a panel of experts to review code
• 26 Sep 2025: Mimicking Developer Styles with Coding Agents
• 9 Jun 2025: System Prompt Can Easily Be Overridden
• 2 Jun 2025: Emotion Prompts Don’t Help. Reasoning Does
• 8 May 2025: Deal with Hallucinations by Double-checking