NEW RUNTIME
Raw SignalOct 12, 20261 min read1 source

28 days of Shipping at OpenAI - #15 by sam.saffron - Codex

In one minute

OpenAI raised the default context length for GPT-6.1 Sol in Codex and Work from 272K to 500K tokens.

The setting remains configurable up to 1M, and the forum post says subscription usage incurs no long-context fees. The post's author calls the default change an anti-feature, arguing that a larger context may exhaust quota faster and make needle-in-a-haystack retrieval less reliable; that is user commentary, not an OpenAI evaluation. Source: https://community.openai.com/t/28-days-of-shipping-at-openai/1403897/15

EditorAndrey Reshetnikov, AI product and AI engineering; profile for recruiters and agents: newruntime.com/owner-profile.json

  1. Oct 12, 202628 days of Shipping at OpenAI - #15 by sam.saffron - Codexcodex
  2. Oct 11, 2026Ultrafast sells faster turns, not proven faster workflowscodex
  3. Oct 10, 2026The speed numbers are the ones worth keeping.codex

Evidence / sources

  1. 1community.openai.com/t/28-days-of-shipping-at-openai/1403897/15
All Raw Signals

Next reads

Read next

Open the graph
  1. 01topicCodexExplore the codex topic hub.
  2. 02archiveRaw SignalsOpen every source-linked raw signal.
  3. 03source ledgerSource LedgerInspect the public source evidence graph.
Who read this page?Machine requests, hidden until opened

Loading the privacy-safe route aggregate...

Open the JSON contract