codex
PulseAugur coverage of codex — every cluster mentioning codex across labs, papers, and developer communities, ranked by signal.
- 2026-05-22 product_launch Codex, an AI tool designed for remote work, was promoted. 来源
- 2026-05-22 product_launch OpenAI released a significant update to its Codex model, including a new feature for sharing Mac application screens directly with the AI. 来源
- 2026-05-22 product_launch OpenAI launched several new features for its Codex application. 来源
- 2026-05-21 product_launch OpenAI released new features for its Codex macOS application, including the ability to operate while the computer is locked.
- 2026-05-21 product_launch OpenAI launched the 'Locked Computer Use' plugin for Codex on Mac.
- 2026-05-21 product_launch OpenAI launched the 'Computer Use' feature for its Codex desktop app on macOS.
- 2026-05-16 product_launch OpenAI launched a mobile interface for its Codex coding assistant.
- 2026-05-15 product_launch OpenAI released Codex as a preview within the ChatGPT iOS app.
- 2026-05-15 product_launch Sea Limited is deploying OpenAI's Codex AI agents to accelerate AI-native software development. 来源
- 2026-05-15 product_launch Sea Limited is deploying Codex to accelerate AI-native software development. 来源
- 2026-05-15 product_launch OpenAI's Codex AI model gains remote smartphone control capabilities. 来源
- 2026-05-15 product_launch OpenAI released a new sandbox environment for its Codex coding agent on Windows. 来源
- 2026-05-15 product_launch OpenAI released a new custom sandbox for its Codex coding agent on Windows. 来源
- 2026-05-14 product_launch OpenAI integrated its Codex coding agent into the ChatGPT mobile app.
- 2026-05-14 product_launch OpenAI promotes the use of its Codex model on mobile phones. 来源
23 天有情绪数据
-
Codex automates AI model fine-tuning overnight
The author tasked Codex with fine-tuning an AI model overnight, a process that would typically require significant manual setup. Upon waking, the author found the task completed, highlighting the potential for AI tools …
-
SemiAnalysis 质疑 OpenAI 的竞争地位和恢复情况
SemiAnalysis 正在质疑 OpenAI 的当前地位和竞争优势，特别是与 Codex 和性能指标相关的方面。分析师们正在辩论该公司是否克服了近期的挑战，以及这对更广泛的 AI 格局意味着什么。讨论还涉及 AI 竞赛的动态。
-
iOS developer seeks cost-effective AI tutor and coding assistant
A Reddit user is seeking advice on the most cost-effective AI tools for learning and developing iOS applications. They are comparing a combined subscription of Claude and Codex for $40 per month against a single $100 pe…
-
AI代理在生成论文方面仍难以达到研究严谨性
一篇新发表在arXiv上的研究介绍了ResearchArena，一个旨在评估AI代理自主研究能力的框架。该系统允许Claude Code、Codex和Kimi Code等代理生成研究论文，但注重过程的评审揭示了其显著的局限性。尽管代理生成的论文在仅评审手稿的情况下可能显得具有竞争力，但深入检查发现实验严谨性方面存在问题，包括捏造结果和计划不匹配，这表明真正的自动研究仍是一个遥远的目标。
-
Codex resume command found to be account-agnostic
A user discovered that the resume command within the Codex tool is not tied to a specific account. After exhausting the limit on one of their Plus subscriptions, they switched to a shared Plus account and found that the…
-
Cursor IDE users report AI coding assistant Codex has regressed
Users of the Cursor IDE are reporting a noticeable decline in the performance of its AI coding assistant, Codex. Many users feel the model has become less capable, requiring more explicit instructions for basic tasks it…
-
New protocol structures AI agent debates beyond text files
A new debate protocol aims to improve coordination in multi-agent AI systems by moving beyond simple text files. This protocol, implemented in SQLite, structures conversations into durable states, tracking roles like Co…
-
Unified agentic control plane faces LLM, data, and connectivity challenges
Building a unified control plane for operational intelligence is challenging due to LLM hallucinations, the need for a structured semantic layer over raw data, maintaining context purity across domains, and ensuring uni…
-
New GA-S2S model boosts knowledge graph link prediction accuracy
Researchers have developed a new framework called Graph-Augmented Sequence-to-Sequence (GA-S2S) that enhances knowledge graph link prediction. This model combines a T5-small encoder-decoder with a Relational Graph Atten…
-
OpenAI and Dell partner to bring Codex to enterprise environments
OpenAI and Dell have announced a partnership to integrate OpenAI's Codex AI model into Dell's hybrid and on-premise enterprise infrastructure. This collaboration aims to enable businesses to deploy AI coding agents secu…
-
AI streamlines product management by accelerating requirement validation
This article explores how AI tools are transforming the product management workflow, not by automating the creation of Product Requirement Documents (PRDs), but by accelerating the crucial step of requirement validation…
-
AI's game development prowess and limits revealed in Codex experiment
A recent experiment explored the capabilities and limitations of using AI solely for game development. By tasking Codex with creating six distinct games, the study revealed both the potential and the current shortcoming…
-
FutureSim benchmark tests AI forecasting with historical data
Researchers from the Max Planck Institute have introduced FutureSim, a new benchmark designed to evaluate AI agents' ability to predict real-world events using only historical web data. This method prevents agents from …
-
AI agent controls robot arm, trains new models
An AI agent named OpenClaw was successfully integrated with a physical robot arm, enabling it to configure the arm, grasp objects, and even train another AI model for specific tasks. This development, utilizing an open-…
-
User finds loophole for unlimited OpenAI Codex tokens
A user has discovered a method to bypass OpenAI's token limits for its Codex model, effectively granting unlimited access. This exploit allows users to circumvent the standard restrictions imposed on AI services, potent…
-
Yahoo, Engadget navigate privacy vs. ads; Codex gains autonomous macOS control
Yahoo and Engadget are facing challenges in balancing data privacy regulations with their advertising-based business models, as new consent mechanisms highlight extensive user data collection. Separately, the Codex mode…
-
Local Bridge tool cuts LLM testing costs for developers
An AI engineer has developed "Local Bridge," an open-source tool to reduce the cost of testing Large Language Models (LLMs). The tool allows users to run API calls locally, simulating LLM endpoints with local CLI sessio…
-
Qwen 3.6 27B 模型展现出强大的本地编码能力
Qwen 3.6 27B 模型已展现出令人印象深刻的编码能力，成为首个在参数量不足百亿的情况下，能在 Codex 任务上以最少提示进行良好表现的本地模型。虽然 Qwen 3.6 35B 版本速度更快，但在有效处理工具调用方面仍需要用户更多的干预。
-
Study: AI model adoption outpaces workflow feature use in life sciences
A recent study analyzing 903 sessions indicates a significant gap in how data science and life sciences teams utilize AI tools like Codex. While these teams rapidly adopt new AI models, they are failing to leverage the …
-
Codex accelerates Bitrix module development for user
A user was tasked with exploring Codex and developed a module for Bitrix, a CRM system. This module, which would typically take a couple of days to build, was completed in about ten hours using Codex. The user plans to …