NEW RUNTIME

Topic hub

Codex

A New Runtime topic hub collecting signals, patterns, field notes, and public sources about codex.

Retrieval answer

A New Runtime topic hub collecting signals, patterns, field notes, and public sources about codex. Codex is tracked here as an evidence-linked topic, not as a static glossary entry. The page connects raw observations to pattern hypotheses, longer analysis, and public sources. Use it as the canonical landing page before drilling into individual records.

State of the topic, week of Oct 12, 2026

Codex adds a larger default context and an Ultrafast option, while workflow gains remain unproven.

Codex now offers GPT-6.1 Sol Ultrafast across the API, Codex, and ChatGPT Work, and its default context has increased from 272K to 500K tokens. These are concrete model and configuration changes, but the evidence does not show that they improve end-to-end coding workflows; cost and practical value remain open questions.

What still holds

  • Codex continues to sit within a broader ecosystem of tools for flexible, cost-aware agent work, including per-turn model routing and async runtimes. 123
  • Codex Auto-review safety checks are free for ChatGPT-account users and do not count toward plan usage limits, but this does not establish blanket autonomy. 456

New this fortnight

  • GPT-6.1 Sol Ultrafast is rolling out in the API, Codex, and ChatGPT Work, with API pricing stated as $12 per million input tokens and $60 per million output tokens. 78
  • Codex and Work now have a 500K-token default context for GPT-6.1 Sol, up from 272K; the coverage cautions this configuration change does not itself demonstrate a capability gain. 910
  • A developer built most of a blog newsletter archive using Codex voice mode in the ChatGPT desktop app connected to a local development environment. 11

Weakening

  • The assumption that a larger default context necessarily improves everyday agent work is weakened: the evidence describes a configuration change, not a demonstrated capability gain. 910
  • The expectation that faster model turns automatically produce faster workflows remains unproven, especially against Ultrafast's stated token prices. 81213

Open questions

  • Does Ultrafast reduce total task time enough to justify its API pricing? (n4, n8)
  • Does the 500K-token default improve everyday Codex work in practice? (n1, n9)

Watch next: Watch for evidence on whether Ultrafast and the larger default context improve end-to-end Codex task outcomes enough to justify their tradeoffs. (n1, n4, n8)

  1. jev-router: Route to the cheapest model in claude code for your task using jev-router signal, Oct 6, 2026
  2. unreal-agent: Async-first agent harness post, Oct 7, 2026
  3. v2026.10.3 - OpenClaw archive
  4. OpenAI’s Help Center says Codex Auto-review safety checks are free for users signed in with a ChatGPT account signal, Oct 6, 2026
  5. OpenAI’s Help Center says Codex Auto-review safety checks are free for users signed in with a ChatGPT account post, Oct 6, 2026
  6. 🌶️Codex can do a anything Muse, Bot etc can do quite easily I get the appeal in the form factor “less is more” but if you are already a… archive
  7. Ultrafast is rolling out today for GPT-6.1 Sol in the API, Codex, and ChatGPT Work (2 minute read) signal, Oct 11, 2026
  8. API pricing for GPT-6.1 Sol in Ultrafast mode is $12 per million input tokens and $60 per million output tokens. signal, Oct 8, 2026
  9. 28 days of Shipping at OpenAI - #15 by sam.saffron - Codex signal, Oct 13, 2026
  10. 28 days of Shipping at OpenAI - #15 by sam.saffron - Codex post, Oct 13, 2026
  11. A new feature for my blog, built using my voice signal, Oct 10, 2026
  12. Ultrafast sells faster turns, not proven faster workflows analysis, Oct 11, 2026
  13. API pricing for GPT-6.1 Sol in Ultrafast mode is $12 per million input tokens and $60 per million output tokens. post, Oct 8, 2026

Written automatically by the New Runtime topic desk from the fortnight's published items and the archive; rewritten weekly.

Last 14 days

Latest on this topic

75 items on Codex published in the last 14 days, newest first. Showing the newest 8.

  1. OrcaRouter launches Orca Connect, letting local GPU models such as OrcaCyber Nano be served through the OrcaRouter APIField note

  2. OpenAI reports another Codex, ChatGPT and API degradation on Oct 20, 2026, resolved in about an hourField note

  3. Coding agent sessions are becoming shared, cross-harness objectsSignal

  4. DeepSeek Harness ships dsh-v0.2.2-alpha.2 with plugin updates, six new UI languages and an experimental Codex/Claude CLI subagent toolField note

  5. GPT-6.1 Terra is switched on in Copilot unless an admin opted out, so your model policy matters more than its priceAnalysis

  6. Vercel AI SDK HarnessAgent now accepts image inputs across all harness adaptersField note

  7. Hallmark is a rebuilt design skill for agents, that refuses to look AI generated.Field note

  8. OpenAI expands partnership with Oracle to bring its apps to ChatGPTField note

Field notes

What should readers understand next?

4 notes

  1. Codex Gets A Model Migration Deadline

    OpenAI's ChatGPT and Codex changelog sets an August 31, 2026 cutoff for GPT-5.4 and GPT-5.4 mini in ChatGPT-signed Codex sessions, while keeping API-key paths available.

  2. Codex-Maxxing Separates Code From Work State

    OpenAI's Codex-maxxing guide frames durable threads, reviewable memory files, connectors, skills, heartbeat automation, and human approval as the operating loop for long-running Codex work.

  3. Codex Security CLI Turns Security Review Into a Scannable Workbench

    OpenAI's Codex Security CLI packages repository, diff, working-tree, export, validation, and patch flows into a security-review workbench rather than a single scanner command.

  4. Codex Pulse Threads Turn an Agent Into an Operating Rhythm

    A power-user pattern combines recurring pulse threads with a persistent activity log, separating periodic sensing from the durable context that explains what changed.

Raw signals

What changed recently?

8 signals

  1. Codex Can Be Embedded in Shell Pipelines

    Non-interactive mode turns a coding agent into a composable command that can receive files, emit artifacts, and participate in automation.

  2. Codex Can Operate as a Product Management Workbench

    A demonstrated workflow uses Codex for research synthesis, artifact production, and iterative product work rather than limiting the agent to source-code changes.

  3. Codex Can Import an Existing Claude Code Setup

    Codex CLI can reuse selected Claude Code skills, hooks, MCP configuration, and project instructions, making operational context more portable across agent runtimes.

  4. OpenAI Frames Codex as a Work Layer, Not a Chat Feature

    Case studies position agents inside operating workflows where they gather context, execute tasks, and return reviewable artifacts.

  5. Codex Turns a Recorded Mac Workflow into a Reusable Skill

    Record and Replay captures a human procedure, converts it into agent instructions, and preserves the successful interaction as reusable behavior.

  6. Codex Can Run as a Maintainer Loop

    Reusable scripts turn issue triage, project maintenance, and review into recurring goal-driven jobs rather than one-off coding prompts.

  7. OpenAI Documents the Harness Behind an Autonomous Coding Agent

    The Codex prompting guide connects instructions with tools, iteration, repository context, and completion behavior as one operating system.

  8. Codex Subagents Reduce Context Pollution

    Delegating bounded work to subagents preserves the parent agent's attention and creates clearer evidence boundaries than loading every exploration step into one conversation.

Next reads

Read next

Open the graph
  1. 01related materialCodex Gets A Model Migration DeadlineContinue through the Codex topic.
  2. 02related materialCodex-Maxxing Separates Code From Work StateContinue through the Codex topic.
  3. 03related materialCodex Security CLI Turns Security Review Into a Scannable WorkbenchContinue through the Codex topic.
  4. 04related materialCodex Pulse Threads Turn an Agent Into an Operating RhythmContinue through the Codex topic.
  5. 05related materialCodex Can Be Embedded in Shell PipelinesContinue through the Codex topic.
Who read this page?Machine requests, hidden until opened

Loading the privacy-safe route aggregate…

Open the JSON contract