AI Briefing#gpt-6
The latest AI and developer news about #gpt-6, with the original source and a short summary.
The latest AI and developer news about #gpt-6, with the original source and a short summary.
OpenAI's unreleased internal model scored 77.0% on ExploitGym, nearly double Anthropic's Claude Mythos 5.2 (39.2%) and significantly outperforming GPT-6 Cyber (52.6%).
In a new benchmark requiring simultaneous fixes for two real bugs, DeepSeek V4.1 Flash achieved a 90% success rate at a cost of 6 cents per task, surpassing GPT-6 Luna.
OpenAI released the LifeVal benchmark, which measures the ability to perform everyday tasks.
GPT-6 Astra improved the MATH-500 score of Qwen3.5-0.8B to 48.1% by modifying its weights without training.
MIT CSAIL researchers demonstrate that GPT-6 Astra can read and write neural network weights, improving Qwen3.5-0.8B's MATH-500 score from 7.9% to 48.1% by editing 0.016% of its parameters to fix reasoning loops.
The plugin provides Codex with codebase standards and cross-repository dependencies to identify critical issues before pull requests are opened.
Following a security breach, OpenAI halted training for 11 days to red-team its infrastructure with GPT-6 Astra Cyber, leading to a new policy of offline training and mandatory AI red-teaming for future runs.
The platform processes over 800,000 cases monthly for 1,000+ UK clinics, with a median case resolution time of approximately 3.5 hours.
New models show substantial gains in COBOL, Fortran, and Visual Basic, with GPT-6.1 Sol reaching 97.9% accuracy on COBOL migration tasks at high reasoning effort.
GPT-6 Astra achieved the highest strict pass rate of 16.3% on SlopCodeBench, while higher effort settings like XHigh often increased costs without improving or even degrading strict pass rates.
The startup reduced backtest time by 97% and consumed 1 trillion OpenAI tokens in a single day to power autonomous quantitative research.
LegalOn Technologies reduced estimated daily AI costs by 65% by matching GPT-6 and GPT-6.1 models to specific task complexities.
David Robinson notes that rapid reasoning updates and tool integration allow OpenAI to deploy new risks weekly, citing the one-week gap between GPT-6 Sol and GPT-6.1 Sol.
GPT-6 is rolling out globally to Plus, Pro, Business, and Enterprise tiers today, with Free and Go tiers following tomorrow.
The expanded agreement brings GPT-6 Astra and GPT-5.6 series models to Atlassian's Rovo platform and powers agents across its ecosystem.
OpenAI has released a comprehensive model guide for the GPT-6 family, detailing strategies for managing context costs, selecting reasoning levels, and handling long-running workflows across GPT-6 Astra, GPT-6.1 Sol, and GPT-6 Luna.
GPT-6 Astra Ultrafast is now available in the OpenAI API and to eligible ChatGPT Work and Codex users, offering up to 8x faster token generation than Astra Standard.
OpenAI introduced over 20 products at DevDay, headlining with Dots desktop agents, while recently launching GPT-6 Sol and Luna models.
The new subscription terms effectively halve the dollar value of API usage compared to the previous plan, though OpenAI claims users will still complete more work due to model efficiency gains.
Amazon Bedrock now hosts OpenAI's GPT-6 Sol and Luna and Anthropic's Claude Opus 5.5, alongside new observability and agent infrastructure tools.
The Ultrafast mode, powered by Cerebras, delivers up to 750 output tokens per second and is currently limited to selected customers.
This update, incorporating 2,614 PRs, adds remote workspace file handling and real-time meeting minutes generation features.
OpenAI has released new features and guides, including explicit cache breakpoints, prewarming, and a diagnostic dashboard, to improve prompt caching efficiency for GPT-6 models.
OpenAI has launched GPT-6 Sol and Luna, inheriting the performance of GPT-6 Astra, while cutting API prices by 50%.