model roundup

GPT 4

4 items · started 2026-05-09 · closed 2026-05-13

  1. thariq from the claude code team basically said markdown is a gpt-4 era habit. back when tokens were expensive and context windows were tiny.

  2. need ur expertise here guys. is it confusing how powerful models like GPT 4 sometimes feel like people with amnesia i mean, the moment they hit a real world workflow.

  3. Paper: https://palisaderesearch.org/assets/reports/self-replication.pdf The paper basically shows that some top AI models can create working copies of themselves when given the right instructions. The models figured out how to copy their o…

  4. Full disclosure: I'm one of the founders of Dolly (https://getdolly.ai). Sharing what we actually built and learned.

← all threads