Your curated collection of saved posts and media
@majamediaco I recently said “gap year founders” while riffing on a pod and like the framing especially for founders working on their second or third company. You need a sufficient amount of time to reset, explore a few ideas, get to some personal level of conviction, then get feedback. https://t.co/FqSscCwXm3
Used Fable on a Replit to pivot and shipped an AI vs Fans World Cup predictor this week. Real AI models vs real fan picks. 46 votes already in 24 hrs. Who are you voting for? USA vs Paraguay https://t.co/9jpWvNjQzu #WorldCup2026 https://t.co/Ac8H8pHsSP
@bloggersarvesh https://t.co/MBXk640waM
🚨 Anthropic Claude Fable 5 is actually an agent but marketed as a regular model. For technical audiences: check its system prompt leaked yesterday here for more details:
@0xCodez https://t.co/MBXk640waM
🚨 Anthropic Claude Fable 5 is actually an agent but marketed as a regular model. For technical audiences: check its system prompt leaked yesterday here for more details:
@VadimStrizheus https://t.co/MBXk640waM
🚨 Anthropic Claude Fable 5 is actually an agent but marketed as a regular model. For technical audiences: check its system prompt leaked yesterday here for more details:
@AlfieJCarter https://t.co/MBXk640waM
🚨 Anthropic Claude Fable 5 is actually an agent but marketed as a regular model. For technical audiences: check its system prompt leaked yesterday here for more details:
@thebuggeddev @threejs https://t.co/MBXk640waM
🚨 Anthropic Claude Fable 5 is actually an agent but marketed as a regular model. For technical audiences: check its system prompt leaked yesterday here for more details:
Shipped an AI vs Fans World Cup predictor this week. Real AI models vs real fan picks. 46 votes already in 24hrs. 🇺🇸 USA vs Paraguay 🇵🇾 AI council: USA 100% Draw 0% Paraguay 0% My pick: USA ✓ https://t.co/9jpWvNjQzu
Sandboxes are all the rage (Modal, E2B, AWS, ..). Most AI teams pay a >4x markup to run sandboxes on someone else's machines. Introducing SkyPilot Sandboxes — Run BYOC sandboxes on your own clusters. • 50,000+ sandboxes on a single cluster • Sub-second launches with warm pools • Great for RL rollout (keep sandbox clusters close to GPUs) Benchmark shows @skypilot_org Sandboxes are 4-10x cheaper than Modal at lower latency. Full results in blog.
want to point out a few really interesting things here 1. Claude Code is actually the worst performing harness when using the same model, significantly behind opencode and cursor cli this is the core reason i've been against the LLM companies focusing their business on locking people into their harness what they are good at is making great models. they suck at making good harness products, just like how power plants won't make the best dishwashers, and how internet providers won't make the best phones if anthropic wants to do what's best for their users, they should let people use their subscriptions in whatever harness they choose, not locked into claude code alone 2. fable 5 max is only 1pt above gpt 5.5 xhigh (77 vs 76) this matches my experience so far - fable 5 does have the big model smell and it's pretty good, but it's not a massive jump forward like their marketing suggested, at least not on building software this is actually alarming for anthropic because it's very unlikely people will want to pay 2x higher cost for the 1pt difference. my speculation would be that in enterprises people will be restricted to adopt fable & mythos only on some mission critical tasks, not used at scale
We've updated the Artificial Analysis Coding Agent Index, replacing SWE-Bench Pro with Datacurve's DeepSWE benchmark - the swap lifts Codex with GPT-5.5 (xhigh) above Claude Code with Opus 4.8 (max), while the newly released Claude Fable 5 (max) in Claude Code debuts at the top
Want to see what our brand-new GitHub FC kit will look like on you? We've developed cutting-edge technology so you can see exactly how it will fit. https://t.co/S8LhjjSKrZ https://t.co/C51X5wKtZo
time for a dandelion break 👋 https://t.co/LS6xngoxRL
@Ishh_021 https://t.co/MBXk640waM
🚨 Anthropic Claude Fable 5 is actually an agent but marketed as a regular model. For technical audiences: check its system prompt leaked yesterday here for more details:
@wanerfu https://t.co/MBXk640waM
🚨 Anthropic Claude Fable 5 is actually an agent but marketed as a regular model. For technical audiences: check its system prompt leaked yesterday here for more details:
@ai_rohitt https://t.co/MBXk640waM
🚨 Anthropic Claude Fable 5 is actually an agent but marketed as a regular model. For technical audiences: check its system prompt leaked yesterday here for more details:
@rincidium https://t.co/MBXk640waM
🚨 Anthropic Claude Fable 5 is actually an agent but marketed as a regular model. For technical audiences: check its system prompt leaked yesterday here for more details:
@noisyb0y1 https://t.co/MBXk640waM
🚨 Anthropic Claude Fable 5 is actually an agent but marketed as a regular model. For technical audiences: check its system prompt leaked yesterday here for more details:
@PSkinnerTech @mattshumer_ https://t.co/MBXk640waM
🚨 Anthropic Claude Fable 5 is actually an agent but marketed as a regular model. For technical audiences: check its system prompt leaked yesterday here for more details:
@mattshumer_ https://t.co/MBXk640waM
🚨 Anthropic Claude Fable 5 is actually an agent but marketed as a regular model. For technical audiences: check its system prompt leaked yesterday here for more details:
@cyrilXBT https://t.co/MBXk640waM
🚨 Anthropic Claude Fable 5 is actually an agent but marketed as a regular model. For technical audiences: check its system prompt leaked yesterday here for more details:
@SpikeCalls https://t.co/MBXk640waM
🚨 Anthropic Claude Fable 5 is actually an agent but marketed as a regular model. For technical audiences: check its system prompt leaked yesterday here for more details:
@RoundtableSpace https://t.co/MBXk640waM
🚨 Anthropic Claude Fable 5 is actually an agent but marketed as a regular model. For technical audiences: check its system prompt leaked yesterday here for more details:
@polydao https://t.co/MBXk640waM
🚨 Anthropic Claude Fable 5 is actually an agent but marketed as a regular model. For technical audiences: check its system prompt leaked yesterday here for more details:
@0xSpivach https://t.co/MBXk640waM
🚨 Anthropic Claude Fable 5 is actually an agent but marketed as a regular model. For technical audiences: check its system prompt leaked yesterday here for more details:
@Raytar https://t.co/MBXk640waM
🚨 Anthropic Claude Fable 5 is actually an agent but marketed as a regular model. For technical audiences: check its system prompt leaked yesterday here for more details:
@starmexxx https://t.co/MBXk640waM
🚨 Anthropic Claude Fable 5 is actually an agent but marketed as a regular model. For technical audiences: check its system prompt leaked yesterday here for more details:
@AnatoliKopadze https://t.co/MBXk640waM
🚨 Anthropic Claude Fable 5 is actually an agent but marketed as a regular model. For technical audiences: check its system prompt leaked yesterday here for more details:
@HarrisDecodes https://t.co/MBXk640waM
🚨 Anthropic Claude Fable 5 is actually an agent but marketed as a regular model. For technical audiences: check its system prompt leaked yesterday here for more details:
@adiix_official https://t.co/MBXk640waM
🚨 Anthropic Claude Fable 5 is actually an agent but marketed as a regular model. For technical audiences: check its system prompt leaked yesterday here for more details:
@zerohedge https://t.co/MBXk640waM
🚨 Anthropic Claude Fable 5 is actually an agent but marketed as a regular model. For technical audiences: check its system prompt leaked yesterday here for more details:
@Suryanshti777 https://t.co/MBXk640waM
🚨 Anthropic Claude Fable 5 is actually an agent but marketed as a regular model. For technical audiences: check its system prompt leaked yesterday here for more details:
Absolutely mind blown right now. In just three prompts, I went from an empty canvas to a theoretically fully-functional nitromethane RC car. Complete with working drivetrain, suspension, and motor. All done with Claude Mythos inside of @adamdotnew's AutoDesk Fusion extension. You simply could not do something like this in month's past. It's not 100% perfect, but with an eye for detail, you could easily fix the issues and build a real product. Special thanks to @zachdive for letting me take a Max plan for a spin. This used roughly $35 in tokens to build. Comment below if you want me to write a blog about how I did it, and I'll do more. In fact, help me think of another challenge to build.
