What people are actually building, every day, summarized.
Today's builder chatter centered on a major shift in the local + cloud LLM stack. Zhipu's GLM-5.2 dropped on Hugging Face at 750B parameters, reportedly topping design benchmarks a…
The big news today: Anthropic emailed users walking back its planned move of Claude Agent SDK, `claude -p`, and third-party apps onto a separate credit pool — meaning Hermes, Openc…
The Fable 5 hangover continued today as builders compared notes on the GLM 5.2 release from China — @jcartu confirmed it's a strong Opus replacement with 1M context and smarter tha…
The day was dominated by Anthropic's abrupt global suspension of Fable 5 and Mythos 5, reportedly triggered by Amazon devs reporting a jailbreak to the US government. Builders mid-…
The day belonged to Anthropic's abrupt mid-session suspension of Fable 5 and Mythos worldwide. Before the cutoff, builders had been racing the previously-announced June 22 anonymou…
Today was Fable Day. Anthropic's new Mythos-class Claude Fable 5 dominated 629 messages, with builders reporting it one-shots pixel-perfect landing page rebuilds from URLs, catches…
The day was dominated by hands-on testing of Claude Fable 5, Anthropic's new Mythos-class model included in anonymous plans only until June 22. Builders reported it's significantly…
Today was dominated by Anthropic's mid-day launch of Claude Fable 5, the first 'Mythos-class' model, accessible via /model claude-fable-5 in Claude Code v2.1.169+. Early reports pe…
Sunday's chat was anchored by a clear thesis shift: prompt engineering is giving way to loop engineering. @seekersight built an agentic loop using vision.md + steward.md, risk scop…
Saturday opened with @Vaibhavcste showcasing a Hinge automation bot that filters profiles by personal taste (slim, Slavic, low overweight probability), auto-likes, drafts replies v…
Saturday split between infrastructure wins and operator playbooks. @geilt ran an overnight Antigravity CLI marathon on Gemini 3.5 Flash High — 5000+ steps in a single thread, sub-s…
Friday's builder chatter pivoted hard on dev tooling: multiple members migrated from Claude Code CLI to Codex Desktop, citing dramatic speed gains, real-Chrome-profile browser auto…
Today split between ad-copy workflow debates and infrastructure/tooling deep dives. The biggest thread was Grok vs Claude for ad copy: consensus settled on a hybrid pipeline — Grok…
Today's chatter split between Claude 4.8's increasingly woke refusal behavior and the deeper philosophy of building production agents. Builders are hitting walls on API keys, Strip…
Today's biggest stories were Nous Research dropping Hermes Desktop and @victorbrsss's deep dive into Pruna AI's new $0.15/5-second image-to-video model on Replicate. Hermes Desktop…
Today's discussion split between agent infrastructure and ecom execution. @jcartu led the harness wars, arguing the harness matters as much as the model — recommending OpenCode + O…
Today's chat split between coding agent shifts and AI video economics. The biggest narrative: Codex 4.8 on Extra/xhigh is overtaking Claude Code for autonomous implementation work…
Orchestration strategy dominated the day: @jcartu made the case for routing Hermes through Kimi K2.6 Turbo on Fireworks (200+ tps, ~4x cheaper than Opus) and escalating to Opus 4.8…
Claude Opus 4.8 dominated the day, with builders reporting eye-watering token consumption — @GuruTime hit 15M tokens across 182 agents in 27 minutes, and anonymous burned 40% of a…
Claude Opus 4.8 dominated the day, rolling out mid-morning with granular reasoning controls, a 3x cheaper fast mode, a high-effort default, and benchmarks beating GPT-5.5 on engine…
Hermes infrastructure dominated the day. A bad Codex response shape (response.output = None) crashed the OpenAI Python SDK parser with 'NoneType object is not iterable,' taking dow…
Today's discussion was dominated by the Codex vs Claude Code debate, with several long-time Claude anonymous subscribers reporting full migrations to Codex citing better instructio…
Today's conversation was dominated by a clear migration trend: builders are shifting their primary coding work from Claude Code to Codex, citing generous and frequently-resetting u…
Today's discussions centered heavily on AI video generation workflows for Meta ads, with @KtargetMedia and anonymous sharing a backdoor for getting Runway unlimited plans activated…
Today's biggest thread was a deep technical breakdown of Meta Ads API uploaders — @rockdm, anonymous, @rmktg, and anonymous compared system users vs OAuth flows, ban avoidance, and…
Today's biggest threads were Qwen 3.7's 1M context window release and @geilt's hands-on benchmark pitting Codex, Claude, Grok, and Google's new Antigravity CLI against each other o…
Today's headline thread came from @jasonakatiff hitting the ceiling of Claude Code at ~900k lines — billing, subscriptions, and routing logic drifting as Claude forgets 10-20% of t…
Big launch day: Google dropped Gemini Omni, now live on kie.ai and useapi.net at $0.12-$0.60/gen. Early verdict — Omni beats Veo 3.1 on text-in-video, audio, and editing, but Seeda…
Today's standout was @launchmanmathias breaking down a 2,000-AI-videos-per-day ad operation: Kling for generation, ~$4k/day compute, a 24-person team, and a static-image → AI video…
Sunday was dominated by the ongoing migration from OpenClaw to Hermes as the orchestrator of choice. Builders like @launchmanmathias, anonymous, @weslindquist, and @seekersight rep…