AUTOMATER NEWSROOM

What’s changing.
Why it matters.

New tools, useful workflows, and consequential AI developments.

Subscribe via RSS ↗

Latest briefing

Oct 27, 2026, 1:43 AM EDT
GitHub Copilot Assisted Permissions Is GA: A Model Now Pre-Approves Agent Commands, Paths and URLs
Copilot CLI, the Copilot app and VS Code can now let a large language model approve clearly safe agent requests and send the rest to you. Here is how to turn it on, what admins can lock down, and why it is not a sandbox.
Oct 26, 2026, 11:57 PM EDT
Cursor Adds DeepSeek V4.1 Flash, Putting Every Effort Tier on Its Cost Frontier at $0.60-$0.98 per Task
Cursor's model picker now includes DeepSeek V4.1 Flash with no BYOK proxy required. CursorBench 4.0 data shows all four effort tiers beat every other benchmarked model at the same or lower price per task.
Oct 26, 2026, 10:56 PM EDT
GPT-6.1 Terra Lands on Amazon Bedrock and GitHub Copilot at Half Price
OpenAI's GPT-6.1 Terra arrives simultaneously on Amazon Bedrock and GitHub Copilot, offering a 1M-token context window, 3x faster task execution, and 50% lower token pricing for daily agentic workflows.

Browse all →

Diagram showing Nanbeige4.3-33B's open weights evaluated inside an existing agent harness, with Terminus 2 producing its 80.9 Terminal-Bench 2.1 score, vLLM and SGLang labeled as separate serving engines, and a planned smaller Nanbeige4.3-3B model shown as a future distillation target.

October 26, 2026

Nanbeige4.3-33B: An Apache 2.0 Open Model Claims the Agentic-Coding Lead in the ~30B Class

Nanbeige4.3-33B is a 33B dense, Apache 2.0 model with 512K context that claims 17 of 20 wins in a mixed coding, cowork and reasoning benchmark suite against three ~30B open rivals.

Our AI-assisted editorial method · Explore our in-depth Intel guides