From the archive
Tuesday, 10 November
Folded from 23 sources that day.
Sorted by quality and by how many sources covered it, the same for everyone.
-
Gemini 4 Carbon is now available
Gemini 4 Carbon is now generally available via the API at $2 per million input and $10 per million output tokens. It ranks first on several agentic coding benchmarks, and the /swarm preview in Antigravity coordinates teams of agents.
-
How GitHub Used AI to Port Copilot From TypeScript to Rust in Record Time
GitHub rewrote the Copilot runtime, about 800K lines, from TypeScript to Rust using the Copilot app and agents. One developer finished in a few months what would have taken a team years.
-
Designing for Agents — Zeno Rocha, Resend
Resend's CEO explains why coding agents keep recommending Resend, citing a study where Claude Code picked it first in most cases. He also describes making pricing and the MCP server (now 80+ tools) usable by agents.
-
Issue #760: Type Checkers, Claude Code Hooks, MarkItDown, and More (2026-11-10)
A weekly Python roundup covering how Astral's ty type checker detects unused variables, a guide to Claude Code hooks, and MarkItDown for converting documents to Markdown. It also lists Django security releases 6.1.3, 6.0.10 and 5.2.19.
-
NaiveAI Promises 2,100 Tokens per Second. How Well Does It Code?
Naive-N0.5-Flash is an open-weights coding model derived from Xiaomi's MiMo-V2.5. In the authors' tests it was faster overall than DeepSeek V4.1 Flash and GLM 5.3 Flash, but slower on hard problems and slightly costlier per question.