Codex
Source reports covering Codex. Stories will appear when analysis and community reactions are ready.
26 reports · page 1 of 1
Connected topics are temporarily unavailable. You can still browse stories below.
Source reports
Codex Desktop continuously generates Crashpad pending dumps
tool
24 min read
Users report that Codex Desktop and the unified ChatGPT/Codex macOS app can continuously accumulate Crashpad crash-report files in local pending directories while the app remains usable. The files appear as paired .dmp minidumps and _sidecar.json metadata under ~/Library/Application Support/com.openai.codex/web/Crashpad/pending in older installs and ~/Library/Application Support/Codex/Crashpad/pending in newer ones. Reported storage consumption ranges from 4.9 GB and 54,504 files in a day to more than 300 GB across both paths, with millions of files and little or no warning before disks approach capacity. Several users identified repeated crashes of the Electron/Chromium browser_crashpad_handler helper. Dumps cited a fatal mach_port_request_notification failure, KERN_INVALID_CAPABILITY, followed by EXC_BREAKPOINT or SIGTRAP reports. One reporter said the handler’s self-monitoring configuration causes a new dump whenever the handler itself fails, producing roughly one dump per second, or about 15 GB daily. Other reports describe large inactive backlogs, indicating that missing retention or pruning can also consume substantial storage without an active crash loop. Deleting pending reports recovers space but may not prevent regrowth. Users have temporarily contained growth by periodically deleting reports or making Crashpad directories read-only, although those measures prevent crash-report writing. Reports call for a handler fix plus built-in size, age, or count limits for pending reports.
OpenAI restores 5-hour Codex and Work limits for ChatGPT Plus users
news
2 min read
OpenAI will restore a five-hour usage limit for ChatGPT Plus subscribers using Codex and ChatGPT Work on August 25. The limit returns after several weeks in which Plus users faced only a weekly usage cap across the two now-unified platforms. Users who reach either the five-hour or weekly limit can wait for the relevant usage cycle to reset or buy additional credits. OpenAI may also issue free limit resets, and users can bank resets through some promotions and referral offers. The company reset weekly usage early several times during the temporary change, including after the platforms added another million active users. Thibault Sottiaux, OpenAI’s engineering lead for Codex and ChatGPT, said the five-hour cap helps smooth demand on the company’s compute capacity while preserving relatively generous weekly allowances. He also said Plus subscribers, whom he characterized as relatively casual and new users, can accidentally consume an entire week’s allowance and find the resulting restriction confusing. The five-hour limit will remain disabled for the coming months on the $100 and $200 Pro subscriptions. Sottiaux did not announce changes to Enterprise or Edu accounts, which use a separate credit-based usage system.
Quick impressions: A week of using Codex more than Claude
opinion
3 min read
A developer who used OpenAI Codex more heavily than Anthropic Claude for a week found Codex generally more contained, technical in tone, and less prone to adding architectural complexity. In Ruby and Ruby on Rails changes, Codex produced fewer comments, while Claude more often introduced abstractions, concepts, Sorbet signatures, and type aliases. When both tools implemented the same requirement from the same documents, Claude produced somewhat more complex code but handled additional cases, according to the developer. The developer felt Codex made initial changes faster, but its extensive test reruns and review work eliminated any overall time advantage in finishing pull requests. Codex also encouraged smaller, focused sessions rather than long-running conversations. Claude remained the preferred tool for urgent debugging because of familiarity, not because it was judged superior. Codex made branch-management errors, including rebasing a branch onto main rather than its target branch and creating a pull request with more than 4,000 additions. Codex was also less effective in the developer’s Jira and Atlassian CLI-based environment, while its MCP CLI login flow was preferred because it explicitly requested the `codex mcp login` command and opened the appropriate authorization process. The developer characterized Claude as more likely to anticipate and act on inferred needs, while Codex more closely follows explicit instructions and stops once it appears finished.
Codex on AWS bedrock bug causing 10x charges
tool
2 min read
A GitHub issue reports that Codex CLI 0.147.0 using its native Amazon Bedrock provider cannot enable explicit prompt caching for the openai.gpt-5.6-sol model through the Bedrock Mantle Responses API in us-east-1. The reporter says Codex emits a session-scoped prompt_cache_key, but its HTTP and WebSocket Responses request types do not support prompt_cache_options or prompt_cache_breakpoint fields. The built-in Bedrock provider configuration supports transport and authentication settings rather than request-body transformations, preventing configuration through config.toml. For completed days from August 5 through August 8, 2026, the reporter estimated 3,656 requests generated 171.94 million cache-write tokens, costing $1,182.09 of an estimated $1,386.46 total model spend. Cache writes represented about 85% of the estimate, which was derived from AWS Cost Explorer quantities and Bedrock rate-card pricing rather than finalized invoice amounts. A local session recorded 76 Sol requests, 6.709 million cache-write input tokens, no cached input tokens, and roughly 88,000 cache-write tokens per request; corresponding CloudWatch metrics showed no client errors. The requested changes include support for prompt_cache_options, typed prompt_cache_breakpoint input fields, provider and model capability checks, placement after stable instruction and tool prefixes, and per-turn cache-read and cache-write telemetry.
Auto-research with codex: How I achieved a 232x Faster Kernel
research
26 min read
Codex was utilized to achieve a 232x faster kernel for the qr_v2 problem in GPU mode. The implementation involved optimizing the blocked Householder algorithm and introducing idea diversity to enhance performance.
ChatGPT Desktop (Codex Desktop) for Linux
tool
2 min read
Codex in ChatGPT enables AI coding agents to perform tasks such as building features, complex refactors, and migrations using OpenAI's advanced coding models. It supports multi-agent workflows with built-in worktrees and cloud environments, allowing agents to work in parallel across projects.
OpenAI says my prepaid credits were consumed, refuses to show any record
opinion
3 min read
A customer experienced a sudden loss of 491.8 Codex credits, which vanished without any recorded usage. After purchasing an additional 1,000 credits for $48.40, those credits also failed to appear in the account.
Codex Security
tool
1 min read
OpenAI Codex Security provides a CLI and TypeScript SDK for identifying, validating, and fixing security vulnerabilities in code. It supports scanning repositories, reviewing changes, tracking findings over time, and running security checks in continuous integration environments, requiring Node.js 22 or later and Python 3.10 or later.
Codex is wearing out our devices
opinion
1 min read
Codex is reportedly causing excessive wear on devices due to its resource-intensive operations. Users have expressed concerns about the longevity and performance of their hardware when running Codex.
OpenAI reduces Codex Model Context Size from 372k to 272k
tool
1 min read
Backporting refreshed bundled model metadata to version 0.144 is proposed in a pull request. Suggestions cannot be applied while the pull request is closed, and multiple suggestions cannot be batched if they involve deleted lines or subsets of changes.
Codex Resets
tool
9 min read
OpenAI's Codex has experienced a total of 35 usage limit resets, with an average interval of 8.9 days between resets and the longest gap recorded at 67.7 days. Reset announcements are tracked daily over the last 26 weeks, providing updates for paid users of Codex and ChatGPT Work.
Codex Micro
tool
1 min read
The kbd-1.0-codex-micro, designed in collaboration with Work Louder, enhances agent workspaces with live RGB feedback and tactile controls for frequently used Codex actions. The keyboard includes a Codex Icon Keyset with 32 extra keycaps and comes with warranty and support.
Codex starts encrypting prompts, uses ciphertext for inference instead
tool
2 min read
Encrypted MultiAgentV2 messages in Codex CLI versions post-0.137.0 eliminate a readable task audit trail. This change results from the implementation of message payload encryption merged on June 5, 2026.
I think I have LLM burnout
opinion
2 min read
Frequent interaction with LLMs can lead to burnout, even with average usage rates. Users may engage with models like Claude Code and Codex for coding tasks while actively reviewing and revising outputs.
GPT-5.6 Sol Ultra will be in Codex
news
1 min read
@haider1 Ultra will be in codex. Tibo@thsottiaux返信先: @haider1さんUltra will be in codex.0:54 · 2026年7月6日15.4万再生数89891051052.2K2.2K8282既読: 89 件の返信
GPT-5.5 Codex reasoning-token clustering may be leading to degraded performance
research
3 min read
GPT-5.5 Codex exhibits a pattern where reasoning output tokens cluster at 516, 1034, and 1552. This clustering may correlate with degraded performance on complex tasks.
A way to exclude sensitive files issue still open for OpenAI Codex
tool
1 min read
A mechanism is proposed to mark sensitive files or paths that should not be read or sent to the model, applicable at both repository and global levels. This feature aims to enhance privacy and security in AI interactions with code.
OpenAI frontier models and Codex are now available on AWS
tool
2 min read
OpenAI frontier models and Codex are now generally available on AWS, allowing millions of AWS customers to integrate OpenAI's capabilities into their existing workflows. This availability facilitates the adoption of AI in enterprises by streamlining the deployment of frontier AI models.
Workspace Agents in ChatGPT
tool
7 min read
Workspace agents in ChatGPT allow teams to create shared AI agents that manage complex tasks and long-running workflows within organizational permissions. Powered by Codex, these agents can perform various work-related tasks, including report preparation, code writing, and message responses.
Codex for Almost Everything
tool
4 min read
Codex has been updated to enhance its capabilities for over 3 million weekly developers, enabling it to operate computers, interact with various tools and apps, generate images, and remember user preferences. The update also includes improved support for developer workflows, allowing Codex to handle ongoing and repeatable tasks.
Codex is switching to API pricing based usage for all users
tool
5 min read
Codex pricing has been updated as of April 2, 2026, to align with API token usage rather than per-message pricing. The new credit rates apply to Plus, Pro, Business, and Enterprise/Edu plans.
Enabling Codex to Analyze Two Decades of Hacker News Data
tool
1 min read
Modolap enables Codex to analyze approximately 10GB of Hacker News data stored in parquet files. The initial analysis focuses on comparing mention history of programming languages, specifically Rust versus Go, and MySQL versus Postgres.
LLMs can be exhausting
opinion
5 min read
Working with language models like Claude and Codex can be mentally exhausting, often requiring long sessions to navigate complex options. Users may experience challenges such as context rot and model bloat, leading to a need for breaks to regain clarity and find effective solutions.
GPT‑5.3‑Codex‑Spark
tool
5 min read
GPT-5.3-Codex-Spark is a new model optimized for real-time coding, capable of processing over 1000 tokens per second on ultra-low latency hardware. This model is a smaller version of GPT-5.3-Codex and represents the first milestone in a partnership with Cerebras.
GPT-5.3-Codex being routed to GPT-5.2
tool
1 min read
GPT-5.3-Codex is being routed to GPT-5.2 in Codex CLI version 0.99.0-alpha.10. The user is subscribed to ChatGPT Pro and is operating on a Darwin 25.2.0 arm64 arm platform using the Ghostt terminal emulator.
The Codex App
8 min read
The Codex app for macOS allows users to manage multiple agents, run tasks in parallel, and collaborate on long-running projects. Codex is available for free with ChatGPT Free and Go, and rate limits are doubled for Plus, Pro, Business, Enterprise, and Edu plans.