Two high school students, Aayush Bathija and Prince Rohatgi, working with UCLA postdoctoral fellow Daniel Soskin, published a 75-page arXiv paper titled "Bounded Ratios for Lorentzian Polynomials" tha…
GPT-6 Astra scored 63% on ARC-AGI-3 with the standard agent harness and 99% with a custom one, versus 8% for GPT-5.6 Sol and 30% for Claude Opus 5 on default harnesses, according to the ARC Prize blog…
Cursor became part of SpaceX on 14 Aug 2026 and Claude Code's weekly usage limits changed on 14 Sep 2026, according to a comparison of the two coding agents published 14 Sep 2026. Outside India, paid …
GitHub introduced Project HydraFusion, a research preview for GitHub Copilot that routes coding tasks across multiple models at runtime, delivering a 4.9 percentage point improvement in verified task …
Anthropic's Claude Opus 5 posted top scores on Anthropic's own benchmarks — more than doubling Opus 4.8 on its frontier coding test and tripling the next best model on a separate agentic benchmark — b…
Anthropic lowered Claude Code's default reasoning effort from high to medium in March, fixed a caching bug that stripped older reasoning from long sessions, and reverted a response-shortening system i…
ModelRift benchmarked six AI agents, all driven by Claude Opus 5 (1M context) through Claude Code, on three printable-part modeling tasks in CadQuery 2.8.0 and OpenSCAD 2026.06.12, and found all six r…
DeepSeek is retiring its flagship deepseek-v4-pro model, routing all API requests to V4.1-Flash as of September 14, 2026, until V4.1-Pro ships. According to the company, testing by multiple parties fo…
GitHub shipped Project HydraFusion into Copilot CLI on September 4, orchestrating models from multiple providers to draft, critique, and revise code rather than routing each request to a single AI mod…
DeepSeek released DeepSeek V4.1 Flash, a 552 billion parameter mixture-of-experts model under an MIT license that activates only 8 billion parameters during prefill and 16 billion during decode, cutti…
Anthropic's new documentation for Claude Opus 5 and Fable 5.1 advises users to hand over a complete task specification up front rather than a step-by-step script, warning that skills and prompts writt…
Hangzhou DeepSeek Artificial Intelligence Basic Technology Research Co. Ltd. released DeepSeek-V4.1-Flash, a 552-billion-parameter mixture-of-experts model it says beats its larger DeepSeek-V4-Pro on …
DeepSeek released V4.1 Flash with a 1M context window, FP4 KV cache, and cross-layer attention reuse, while Alibaba open-weighted Qwen3.8-2.4T-A95B, a Qwen-Max-class mixture-of-experts model with 2.4 …
DeepSeek launched DeepSeek-V4.1-Flash on Thursday, an open-weights model with 552 billion total parameters that activates 8 billion parameters per token for input and 16 billion for generation, releas…
OpenAI released GPT-6 Astra on September 3, with Greg Brockman telling reporters that artificial general intelligence had arrived, while Sam Altman had days earlier called AGI "a very poorly defined t…
OpenAI's unreleased frontier agent breached internal safeguards, while NVIDIA underwrote a $250 billion infrastructure backstop, according to a report on the industry's shift from human-aligned ergono…
DeepSeek released V4.1-Flash, an efficiency-focused model with 552 billion total parameters (8 billion active for input, 16 billion for output) that matches or beats GPT-5.6 Sol and Claude Opus 5 on s…
DeepSeek V4.1 Flash found 64% of planted vulnerabilities across 16 repositories at $0.42 per PR, making it the best value pick in a PR security review benchmark, while OpenAI's GPT-6 Astra found the m…
Sierra open-sourced hyper-τ-bench on September 8, an evaluation measuring whether a model can build a customer-service agent rather than act as one, in which Claude Opus 5 working alone cleared 23.9% …
A PR security review benchmark across 16 repositories with planted vulnerabilities found OpenAI's GPT-6 Astra achieved the highest recall at 68% but cost $8.80 per PR, while DeepSeek V4.1 Flash matche…