AINewsnow

27,280 Agent Runs Prove Your Harness Matters More Than Your Model

27,280 agent runs. 83 billion tokens. 22 harness configurations, 5 models, 62 terminal tasks, 4 attempts each. The headline is not which model won. It is that the same harness moved eleven of twenty-two places when the model underneath it changed. Codex CLI ranked first on cost per task with GLM-5.…

Read the full story at DEV Community — AI ↗

Timeline · 1 report

  1. 2026-11-18 00:34 · DEV Community — AI
    27,280 Agent Runs Prove Your Harness Matters More Than Your Model

More stories

  1. Baseten Inference Infrastructure Enters Public Preview on Microsoft Foundry — Unite.AI
  2. China AI Highflier Crashes 85% as Chip Fears Hit Model Makers — Bloomberg AI
  3. Alibaba's QwenWork Adds Third-Party Models From Kimi, GLM and DeepSeek — Pandaily
  4. OpenAI launches GPT-6.2 Astra and Sol as Google ships Gemini 4 Neon and Microsoft unveils MAI Code 2 — OpenAI News
  5. Anthropic debuts on Nasdaq with record $81.3 billion IPO, hitting $2 trillion value — Anthropic (X)
  6. Google launches Gemini 4 Neon and opens Genie 4 world model to Gemini API developers — Google DeepMind YouTube
  7. Google books Blue Origin's New Glenn for 2029 launch of AI server satellites — Google: The Keyword
  8. AWS expands SageMaker AI with data preparation tools and cross-Region inference routing — AWS Machine Learning Blog

Read the Daily Brief: the day's most important AI stories →