#编码

Your Site is live. And now you can see how it’s doing. Sites Analytics is in public testing, giving you basic performance metrics for the sites you publish with Codex.

Ling-3.0-flash is now free on OpenCode inclusionAI's latest model optimized for token efficiency

The gap between flash and frontier models keeps shrinking. Ling 3.0 Flash is live on Kilo now, free for a limited time, and it's punching way above its weight for agentic coding tasks. Big thanks to @NovitaAI for the speed behind it!

gpt 5.6 pro is substantially smarter than 5.6 in codex. it is inordinately good at critiquing and fixing 5.6's dumb ideas and implementations. there must be a better way than asking 5.6 codex to write research reports, me pasting into pro, then copying instructions back!

Your coding agent can now process documents without leaving the IDE. Add Agentic Document Extraction (ADE) skills to Claude Code, and the setup takes two commands. Connect to the plugin marketplace, install the ADE Document Processing skill, then reload your plugins. Once the skill is active, your agent parses documents into structured Markdown and pulls structured fields directly inside your workflow. No switching tools, no writing client boilerplate from scratch. Full setup steps live at https://docs.landing.ai/ade/build-with-ai-agents#skills.

Deepdive: Kimi K3 max vs. GPT 5.6 Sol max on software engineering/DeepSWE tasks. Both of these models provide nice specializations and routing/cascading between them is the right way to go! full deep-dive 👇(1/n)🧵

http://x.com/i/article/2079482473818333184

Keep work across multiple folders in one Codex project. Local projects can now include related code, docs, and reference files from multiple folders. Codex can read and write across them while one primary folder remains the Git root.

already in bed. couldn’t remember if my laptop was plugged in. asked codex. good night

A delivery orchestration platform that supplies FedEx, estimated 7 to 8 months to rebuild their core system. We delivered in 3.5 months with 2 engineers and 122 merged pull requests in the first 90 days. I want to walk through how, because "we used AI and it was faster" doesn't help anyone. Their codebase had real users, real logistics operations, and real consequences if something broke. This is brownfield, not a weekend prototype demo. Week one, we didn't write code. We scanned the entire repo and built a knowledge graph: markdown files documenting every module, dependency, data flow, and domain term. Our agents read that graph before touching anything. Teams skip this constantly. They point an agent at a repo, hand it a ticket, and get code that compiles but misunderstands the architecture. I've seen it across 100+ engagements. When someone says "AI doesn't work on our codebase," they almost always skipped context acquisition. Every ticket ran a six-step loop: define, spec, plan, implement, test, document. The agent handles about 90% of code generation. A senior engineer reviews every PR, and nothing merges unless they can verify it, explain it, and debug it without the agent. We call that the V.U.E. gate. The 50% time reduction came from three things: upfront context eliminated false starts, the agent handled mechanical coding so engineers focused on design and review, and tight spec-to-PR discipline kept scope from drifting. AI compute: about $200 per developer per month. Their CTO called our pod the "top performing team" inside the company. Both engineers got discretionary bonuses, twice. That's what our Velocity Framework was built for.

We've tested the Qwen 3.8 Max preview and it absolutely rocks! Each scene is built in one shot, a single Three.js file with nothing else added: • Odysseus' war galley riding the waves • Trojan horse • Greek war helmet Qwen managed to execute the task faster than the other models, while output quality is relatively the same across all of them. If the pricing stays close to Alibaba's previous flagships, we'll get one of the best speed/quality ratios on the market — perfect for running agents at scale. Soon live on AI/ML API.

People are like "You're just running a lifestyle business." “Yeah I have a business and a lifestyle and my lifestyle happens to include a fucking yellow Lamborghini.”

Grok Build just got major update, giving developers more control over AI tools, better account visibility, built-in diagnostics, stronger session reliability, and lower memory usage for Voice Mode on macOS. Release Notes: v0.2.111 Features: • Users can now disable image generation and video generation tools (and their slash commands) via config.toml or environment variables. • /session-info now displays whether the session uses OAuth or an API key and where to manage the account. • You can now run grok doctor fix commands directly from inside the TUI instead of only from the CLI. Bug Fixes: • !cmd commands now allow up to one hour before timing out. • npm package now installs the native binary under $GROK_HOME/bin (honoring the same override as the Rust CLI). • Startup warnings now point to /doctor for details and fixes. • Dashboard hover and clicks no longer miss the gaps between items in wide mode. • Shift/Alt+Enter now inserts a newline while editing a queued prompt. • Queued prompt edits under combine mode no longer lose changes due to premature hold release. • Forking a session that used compaction no longer causes later rewinds to fail with missing checkpoint errors. • When a permission prompt appears while viewing scrollback, focus now correctly moves to the prompt so you can answer. • Pressing Esc once now cancels the current agent turn (except in fullscreen vim scrollback mode). • Grok now automatically stops a turn that keeps repeating the exact same tool call many times in a row. • Configs using either spelling of the workspace teleport disable flag now load and save correctly. • Background subagent completion messages no longer leak into unrelated sessions when multiple sessions are active. • When the auto-permission classifier times out or fails, Grok now shows a normal permission prompt instead of silently denying. • Managed MCP tools no longer time out prematurely on slow operations like Notion updates. Performance: • Voice dictation on macOS now uses less memory by running capture in a temporary helper process.

Grok Build, with the Browser Use CLI, outperforms Codex and Claude Code at browser tasks

Unbelievable excited for what’s coming together. Tomorrow is feeling codexy

We summarized the key remarks from DeepSeek founder Liang Wenfeng during their investor meeting. The core topics include open-source strategy, cost reduction, and AGI. https://www.thexpin.com/p/the-deepseek-doctrine

Hy3 now available in Go text · 256K context latest model from Tencent