bookmarks
4,281 bookmarks

Introducing SWE-1.7, the most capable model we’ve trained yet. It scores within a few points of the strongest frontier models at a fraction of the cost, and is now available at 1000 tok/s. RL is not hitting its limit: after refining our recipe, we keep seeing gains as we scale↗

·@cognition·Jul 8, 2026·model-release·inference-speed·model·rl·swe-1-7

If you have access to GPT Live, test out this prompt. "Hey, I want you to teach me about a concept. <insert concept> but every time I interrupt you, I want you to get progressively more angry at me and tell me to stop talking and let you finish" There is an amazing feeling w…↗

·@jxnlco·Jul 8, 2026·tool·gpt-live·prompt-engineering

Introducing GPT-Live, a new generation of voice models for natural human-AI interaction. Rolling out in ChatGPT starting today. You’ll want to turn the sound on for this one.↗

·@OpenAI·Jul 8, 2026·tool·chatgpt·gpt-live·openai·voice-model

GPT-live (next-generation voice) launches today in ChatGPT. it feels magical and 'real'. i have always preferred typing to talking to an AI, now i think that's going to shift.↗

·@sama·Jul 8, 2026·media·chatgpt·gpt-live·voice

i have this file to replace the claude code system prompt with a pi-style one alias claude-pi='claude --system-prompt-file ~/.claude/pi-system-prompt.txt --dangerously-skip-permissions'↗

Here's a step-by-step process to kill all the bloat from your Claude Code system prompt: 1. Run a proxy so you can see exactly what gets sent to Claude Code (included in the article) 2. "Fuck, there is so much cruft in there" 3. Use my settings.json to kill all the bloat Down t…↗

That's the spine. Fair hit. That's something to sit with. A real observation. That’s the whole thing. Sharpen that: say the word. Notice the arc of what just happened. One honest caveat: the full amount, stated plainly. Genuinely. Quietly. Honestly. That’s doing real work.↗

Fable is extremely good at writing original NES games in assembly. Here it designed an original game graphics, soundtrack, everything from scratch, and it'd actually run within the constraints of a real cartridge. Wild.↗

·@skirano·Jul 7, 2026·demo·assembly·game-development·nes

Router has been in production for a few weeks and the savings have exceeded the original benchmarks. Measuring billed usage vs the same workload priced at Opus rates, we’ve found: 43% aggregate cost savings 61% of sessions are at least 80% cheaper↗

I'm open-sourcing my Agent Skills library. 75 skills for Codex, Claude Code, Cursor, and other agents, focused on web design, landing pages, motion, WebGL, UI styles, and assets. A few favorites: - Video to Super Prompt Turns a screen recording of a design, landing page, or an…↗

Introducing AI Browser Games! Watch open & closed models build small browser games head to head. Open models like Kimi K2.7 were faster, cheaper, & produced games similar to Opus 4.8. In some cases, open models were ~20x cheaper than closed ones too!↗

Claude Cowork is coming to mobile and web. Hand Claude a task at your desk and pick up the finished work from your phone. Close the laptop and Claude keeps going. Beta is rolling out over the next several weeks starting with the Max plan, with more plans to follow.↗

·@claudeai·Jul 7, 2026·tool·beta·claude-cowork·mobile·web

Don't make the same mistakes I did. Here's a video on how to get the most out of Fable, full of real world examples of Fable helping me ship.↗

·@theo·Jul 7, 2026·media·fable·how-to·real-world-examples

This is an insane Fable 5 UI/UX hack. This "Taste" skill completely kills generic AI-slop and gives Fable 5 the tools & instructions needed to ship beautiful design. This might just be the best AI skill I've ever used. tasteskill.dev↗

·@milesdeutscher·Jul 7, 2026·tool·fable-5·taste-skill·uiux

LLM Wiki's Don't Work You read the Karpathy post. Everyone is talking about LLM Wikis. You want one too. You open Obsidian, run an LLM against your docs, and start pulling company context into one place: strategy docs, meeting notes, sales calls, product decisions, customer thr…↗

Do this on your last day with Fable You can extract Fable 5's intelligence out of the model before it's gone... and i'm going to teach you the 5 workflows that do it, with every prompt ready to paste because tomorrow Fable 5 leaves your subscription it moves to pay-per-token c…↗

·@EXM7777·Jul 7, 2026·workflow·claude-code·claude-md·fable-5·obsidian-vault

Memory cost and capacity are significant issues for AI accelerators. Unlike game rendering, model inference can have a deterministic memory access pattern. You don’t need “random access memory” at all for model weights, and you could tolerate cold-start latencies in the multiple…↗

You will thank yourself for sending this prompt to Fable: "Read this article and let me know what you would recommend doing off the back of this, whether it's sending things to [project]'s codebase to investigate stuff, asking me questions, or whatever else you recommend. Whethe…↗

·@robj3d3·Jul 6, 2026·tool·fable·prompt·reflection