News
Opus 5.1 could be releasing soon
An X post reports that Opus 5.1 correctly recalled the Opus 4.6 release date in a no-tools test, while base Opus 5 reportedly gave less reliable answers.
Prompt Blueprints / Topics
The latest in AI.
Releases, research and the stories shaping AI. Reporting across the companies and models that matter.
More stories
News
Anthropic Tests Autonomous AI Alignment on Small Models
A new research announcement says Claude researched, trained, and tested methods intended to improve the alignment of small models within 48 hours and on one…
AI releases
Hy4 Preview opensource model released
Tencent Hunyuan has announced the Hy4 preview with 770B scale, 49B active, a 1M context, and a call for hands-on feedback.
News
Gemini Omni 1.1 Flash Adds New Video Controls
Google AI Studio’s latest model update adds scene extension, first-and-last-frame control, video references, faster 360p previews, and 1080p or 4K output for…
News
Nvidia-Hugging Face Acquisition: What We Know
Nvidia is reportedly pursuing a roughly $13 billion acquisition of Hugging Face, a major open-source AI platform. The deal remains unconfirmed, but it could…
News
X Bot Accounts: Native Identities and API Controls
The X bot model described in the post gives agents distinct identities, visible ownership, API-managed lifecycle controls, and token security.
News
Claude Brings a Built-In Browser to Cowork
Claude’s Cowork can now open a browser in its side panel for website tasks, allowing it to navigate, fill out forms, and finish the job.
News
GLM-5.3-Flash Scores 57 on Artificial Analysis Index
GLM-5.3-Flash is reported at 57 on the Artificial Analysis Intelligence Index and is shown connected to GLM-5.3 along the Pareto frontier.
News
Introducing Gemini 3.5 Transcribe
A focused news brief on Gemini 3.6 Transcribe's stated ability to handle self-corrections, remove filler words, and produce clean, formatted text for…
News
Fable 5.1 May Arrive Before Astra
A post suggests Fable 5.1 could be released before Astra, reversing an expected order while leaving the timing unconfirmed.
News
Qwen3.8-Flash Announced as Open-Weight Qwen3.8 Flash
Alibaba Qwen has announced Qwen3.8-Flash as an open-weight multimodal MoE model and an early preview of the Qwen4 architecture, with a planned QwenCloud API…
News
ChatGPT Work Begins Rolling Out Website Login Support
ChatGPT Work is rolling out website login support on web and mobile for Plus, Pro, and Business users, allowing them to ask the assistant to handle tasks that…
News
OpenAI Bel Pretrain Reportedly Tops 10T Parameters
A focused report on a claimed OpenAI pretraining milestone and what it could mean for Astra, GPT-6, and the company’s reported competitive position against…
News
Apple M6 and M5 Ultra Launched!
Apple’s M6 and M5 Ultra Launches!
News
ChatGPT Plus 5-Hour Limit Returns for Work and Codex
The five-hour usage limit is returning for Plus accounts across ChatGPT Work and Codex, while the two Pro tiers remain exempt for the upcoming months.
News
GPT-5.6 Sol Pricing Drops to $4/$10 per Million Tokens
GPT-5.6 Sol pricing is reported to have fallen to $4 per million input tokens and $10 per million output tokens. This brief explains the reported reductions…
News
Antigravity Remote Control Expands Access to Active Sessions
Antigravity is rolling out Remote Control, giving users access to active sessions through supported browsers and iOS and Android devices, beginning with Ultra…
News
GPT-5.6 Sol Pricing Cut by Over 20%
OpenAI says it will cut GPT-5.6 Sol API and credit pricing by more than 20% for three months, giving users a temporary reduction in usage costs.