Tinker

Most AI accounts either sell hype or talk over your head. I do neither. I test new tools, break things, and tell you what actually works. Not a developer. I figure things out with AI, not around it.

@TeksEdge I don’t want to sound like a douch but this sounds like a shill . I mean… it sure is a great model but come on..

🆕 Fable 5 vs Nex-N2-Pro. Anthropic's new Fable 5 did NOT one-shot my Cosmic Dodge prompt, unlike Nex-N2-Pro. imho Nex-N2's output is more exciting, gameplay more aggressive, and bonus drops faster, but the outputs are still very close. 🏆 I give Nex-N2 the WIN, but very close. ✦ Fable 5 $50/1M > Nex-N2 $0/1M ✦ Nex-N2 1-shot > Fable 5 2-shot

@bridgemindai DAMN LETS GOOOO

Claude Fable 5 is now available in Claude Code.

@bridgemindai Lmao

Using Claude Mythos to center a div

@bridgemindai Let’s go. You gonna stream it?

@tinkerlog Absolutely sure

GPT 5.6 is not going to compete with Claude Mythos. Claude Mythos is a completely new class of model.

@DanielGPT2022 @godofprompt Nice !

@tinkerlog @godofprompt A peptide trust verification app.

The most powerful model Anthropic has ever built is reportedly about to go public. It's called Mythos. Until now, it's been locked to a small security consortium, where it found over 10,000 critical software vulnerabilities in a single month. Alex Heath reports the public version could land as soon as tomorrow, possibly under the name Claude Fable. Anthropic held it back over security risks. If it ships to everyone, the gap between operators who can direct a model like this and everyone else is about to get a lot wider.

@bygregorr @godofprompt Why do you say that?

@godofprompt not sure 10k criticals in a month is the signal it sounds like any scanner set to a loose severity threshold can hit that count. did they share anything on false positive rate or what actually got patched?

@xkaidus @godofprompt You sure ? I think it’s gonna be a half baked one

@godofprompt so theyre finally letting Mythos out of its cage hope the fruitcake devs are ready lmao

@DanielGPT2022 @godofprompt What app ?

@godofprompt Looking forward to finishing my app with it!

@AndreBcker18745 @bridgemindai Codex basicly the same but much less pricer…

@tinkerlog @bridgemindai 100% agreed here. I am excited how they are planning to operate in the future and how their pricing develops. I think that anthropic will still be on top for a little while longer, but once other firms catch up, it Starts to get interesting

@canbaz2023 @kimmonismus Why do you say that ?

@kimmonismus Anthropic ships a model that breaks CVEs in hours. Cool. Meanwhile Sonnet is still the same. Some of us just need a fast, cheap model that thinks clearly. Is that too much to ask?

Claude Mythos will be released today (June 9th), according to leaks everywhere. The interesting question is whether they'll also update Sonnet or Haiku. Not that Mythos isn't enough for me, I'm just curious why the smaller models are currently getting so little attention. Anyway, it will probably be called Claude-5-fable.

@MRehan_5 @kimmonismus Why?

@kimmonismus Anthropic is not releasing yet

@thegreatest_sv @kimmonismus Why?

@kimmonismus I think we’re just gonna get some minor updates on the models we are currently using. Nothing special is gonna happen.

@antideologist @kimmonismus Yeah probably..

@kimmonismus Mythos definitely isn't enough for me because it will be basically unusable in Pro usage limits even once released...

@Werrzz @kimmonismus Why? I find it usefull for basic things

@kimmonismus At this point they should just retire Haiku. Competition is too strong on this low cost tier. Ideally they will make Sonnet faster and tune it for agentic sub tasks and tool usage.

@YuriKushch @claudeai Let’s goooo can’t wait

@claudeai Guys we are waiting for Mythos aka Fable!

Final stop: Tokyo. Register to hear directly from the teams behind Claude: claude.com/code-with-clau…

@AndreBcker18745 @bridgemindai I understand , but they can’t keep raising the price . They are gonna be crashed if they do it . Other models are getting cheaper , not Just the Chinese ones . Only a Matter of time before xai or Gemini get on their same level but with a fraction of the cost .

@tinkerlog @bridgemindai I mean, anthropic is already pretty expensive, now with such a Model coming up and all the Hype, i don‘t think that‘ll be chest tbh.

@AndreBcker18745 @bridgemindai Why do you think it’s gonna be so expensive? That’s only rumors . Lets see what the actual cost will Be

@bridgemindai The only Point where i See 5.6 competing is the Price. Mythos will be insanely expensive and Most people have to use their limits and usage wisely. But mythos will dominate, yeah

Everyone's comparing Ideogram 4.0 to Midjourney on how pretty the images look. Wrong test. The real question for an open image model is whether I can hand the output to a client as a finished design asset. Clean text rendering, transparent backgrounds, layout control by region. That's a production pipeline, not a gallery. Judge it on the work, not the wallpaper.

@Cross_of_cris @astropol0 @elonmusk @grok Completely agree. Thats sad, I also bought the premium + to access it when it was actually great…

@astropol0 And @elonmusk doesn't give a damn. Ever since he started censoring things at the beginning of the year, he's been killing and ruining @grok, plus the pathetic, absurd limit. Now it's just a complete scam. Many thought he cared, but it's lie after lie with what he says every week.

Grok 4.3 is getting cooked on the Agent Arena leaderboard.... It’s literally performing below average in real agentic tasks, -9.5% net improvement, with terrible scores in success rate and steerability. Even Grok’s own Build 0.1 version is beating the main model. This is actually embarrassing @elonmusk please MAKE GROK GREAT AGAIN!