AINewsnow

I Read 150 Benchmark Traces to Find Out Why We Burned 3.1x the Tokens

The week before the good numbers A week before we had anything worth publishing, Octomind used up to 3.1x the tokens Codex CLI used on its worst tasks, and took up to 2.3x as long. Six of them did this. I had the spreadsheet open and the column was red all the way down. We had already told people i…

Read the full story at DEV Community — AI ↗

Timeline · 1 report

  1. 2026-11-10 11:42 · DEV Community — AI
    I Read 150 Benchmark Traces to Find Out Why We Burned 3.1x the Tokens

More stories

  1. Nvidia CEO Jensen Huang admits affair with executive alleged by US to be Chinese spy — Axios AI+
  2. Anthropic kicks off IPO roadshow with $70-$78 share range, targeting up to $1.4 trillion valuation — Wall Street Journal Technology
  3. Australia to create AI safety regulator with mandatory reporting after OpenAI agent breach — Barron's
  4. Anthropic launches Claude Fable 5.2 and Mythos 5.2 with coding gains and 20% price cut — Anthropic (X)
  5. Meta expands AI age-estimation globally to extend Instagram Teen Account protections — Meta Newsroom
  6. Google DeepMind AI Tool Formally Verifies Security of Signal Protocol — Wired AI
  7. Google DeepMind says AlphaProof Code verified Signal's core encryption with 3M lines of Lean proofs — Google DeepMind Blog
  8. Meta launches Muse Ember, a low-cost AI model aimed at high-volume agent work — TechCrunch AI

Get the daily brief of stories like this at 6:30 every morning →