OpenAI proposed a new AI performance metric called 'Useful Intelligence per Dollar' to help enterprises measure the economic value of AI outputs rather than traditional software adoption metrics. The …
OpenAI released GPT-5.6 to the public on July 9 as three tiers—Sol, Terra, and Luna—all running on the same 4T parameter Spud pretrain base, with tiered pricing based on inference compute. The release…
OpenAI's GPT-5.6 Sol model achieved a 46.2 mAP@50 in object detection and 73.0% accuracy in object counting, significantly outperforming its predecessor GPT-5.5, according to benchmarks by Roboflow. T…
OpenAI released GPT-5.6 on July 9 as three distinct models—Sol, Terra, and Luna—with different pricing and capabilities, sharing a 1.05 million-token context window and 128K max output. The default gp…
OpenAI introduced GPT-Red on July 15th, an internal automated red-teaming model designed to find prompt injection vulnerabilities at scale and feed those attacks back into training production models. …
OpenAI released three models named Luna, Terra, and Sol, which correspond to small, medium, and large variants of GPT-5.6. The naming reflects an inference cost architecture based on the roofline mode…
OpenAI CEO Sam Altman announced the GPT-5.6 Sol model, which delivers a 54% improvement in token efficiency for agentic coding tasks, but warned users to expect hiccups and capacity crunches during th…
OpenAI removed the five-hour Codex usage cap, replacing it with a weekly limit for some accounts, and released GPT-5.6 alongside three lower-cost models—Sol, Terra, and Luna—that offer competitive per…
OpenAI released GPT-5.6 on July 9, 2026, as a three-tier model family—Sol, Terra, and Luna—designed for agentic tool calling. All tiers share a 1M-token context window, 128K max output, and native MCP…
Charlie Labs announced that its AI-powered workspace agent Charlie will now enforce daily and weekly Credits limits included with each pricing plan, after previously running without usage charges for …
OpenAI published new prompting guidance on July 9, 2026, for its GPT-5.6 model family, urging developers to use outcome-first prompting instead of step-by-step instructions. The approach reduces token…
OpenAI released GPT-5.6-Sol, along with cheaper models Terra and Luna, positioning Sol as a workhorse for practical tasks like coding and agents while acknowledging Anthropic's Fable remains superior …
OpenAI released three permanent model tiers named Sol, Terra, and Luna under the GPT-5.6 generation on July 9, replacing the previous version-number churn with durable capability tiers. The update int…
OpenAI released GPT-5.6 Luna, Terra, and Sol, which cleared the High cybersecurity capability threshold under its Preparedness Framework, enabling automated end-to-end attacks and vulnerability discov…
A guide explains how to use GPT-5.6 Sol as an orchestrator with cheaper sub-agent models like Luna and Terra to reduce costs by 60-80% while maintaining output quality, emphasizing that most tasks do …
OpenAI released GPT-5.6 with three tiers (Sol, Terra, Luna) and a Sol Ultra mode, offering distinct pricing and performance levels for AI agents. Early production migrations to the mid-tier Terra mode…
OpenAI's Codex team revealed during a Reddit AMA on July 10 that Codex now has over five million weekly users, doubling in three months, and shipped 150 features in that period. The team recommended G…
OpenAI released GPT-5.6 as three tiers—Sol, Terra, and Luna—each with different pricing and performance characteristics. An engineer benchmarked all three on real traffic and found Terra hits about 97…
Artificial Analysis evaluated all 15 combinations of OpenAI's GPT-5.6 Sol, Terra, and Luna models across five reasoning effort levels. The analysis found that Terra is never on the intelligence-vs-cos…
OpenAI launched GPT-5.6 on July 9, introducing a three-tier pricing model with the Luna tier at $1 per million input tokens, a 25x cost reduction compared to competitors. The aggressive pricing target…